Surge suppressors seem to be used all over these days, any opinions on using one in a 928?

*** obviously I don't mean a power strip, but a device targeted to the safe voltage range in our cars.

 

We have lots of them already. They are called "fuses".

 

No I meant surge suppressors that control over voltage situations, fuses are current limiters that often respond to over voltage much slower than even light bulbs let alone electronics.

 

This can be a long discussion, but the short answer is no, you don't need them. Surge suppressors are useful to protect complex electronics from the current surges(not voltage) that occur on startup. Thankfully, the type of products made for the auto market go through a pretty thorough testout and reliability testing in design. Auto electronics will handle a lot of variation in voltage, and still maintain operation.

over voltage protection on most electronics that are DOT certified is good for up to 21 volts for a few seconds, and 32 volts for a few milliseconds. Your charging system will never get anywhere near 21 volts DC, as it is regulated at about 14 volts. Even open circuit with no regulation the alternator will put out no more than about 15.5 volts(depending on make and model).

Batteries act as a huge surge brake for things like voltage and current spikes. Very effective at damping out large spikes and drops.

Doc

 

The car comes with a large low-impedance surge suppressor, located in the rear under the spare tire. If your cables and connections to it from the alternator are good and your ground connections are clean and secure, the group-48 wet-cell surge suppressor that the factory supplied should be fine for any situation a 'normal' 928 should experience.

Note also that some cars have diode suppression in several of the relays in the panel. In my S4, there is a dedicated diode surge suppressor in the X bus for the feed to the AC system. Probably to help control some of the local reverse-EMF from solenoid of compressor clutch coils.


Most 'common' surge supressors you find in power strips and the like are shunt-trip devices that cause an overcurrent event, tripping a thermal breaker when an 'overvoltage' event happens. They are often rated at 150% of the expected system voltage, so probably wouldn't save anything in a digital car. More of the gimic devices sold to 'absorb' voltage spikes in the DC systems are nothing more than capacitors with current-limiting and drain resistors built in. The battery in the car really does a much better job than these ever will.

 

"over voltage protection on most electronics that are DOT certified is good for up to 21 volts for a few seconds, and 32 volts for a few milliseconds"

Do what? And where's this speced, please? Obviously you haven't evaluated many ECU designs.
Ask owners whose LH units failed because of improper battery charging, i.e. greater than 16-18 volts.

"Your charging system will never get anywhere near 21 volts DC, as it is regulated at about 14 volts."

Right! How about when the regulator fails and full-fields the alternator (110 amp Bosch one).
Please stand back!

"Even open circuit with no regulation the alternator will put out no more than about 15.5 volts(depending on make and model"

Come on! This is the real world.

"Batteries act as a huge surge brake for things like voltage and current spikes. Very effective at damping out large spikes and drops."

Right, but only if they're in good condition. Most, though, forget about localize transients which
occur a distance from the battery. Remember, the harnesses do have small amounts of inductance
and resistance which essentially isolates the low impedance of the battery from short term spikes,
e.g. milliseconds.

"More of the gimic devices sold to 'absorb' voltage spikes in the DC systems are nothing more than capacitors with current-limiting and drain resistors built in"

Not really. Most use specially designed semiconductor devices (like a zener diode) which
can absorb short term transients of limited energy expressed in Joules. They come in various
voltages and can be used to protect devices like the original LH ECM.

Bottom line: Most forget that the internal impedance of a battery (thats which affects its
ability to act as voltage limiter) varies with the condition of the battery. Most later ECUs
are designed with internal protection, e.g. series diode (reverse protection) and a transient suppressor.

Read here for more info: www.systemsc.com/problems.htm

 

The surge allowance for auto electronics as mentioned is a DOT spec. You may look it up if you want. I haven't seen it in many years, because I don't work in the auto field anymore. The spec is there for two reasons. First, the voltage fluctuations on cranking then charging is very great, so the power regulation of DOT devices is quite strict. Next, the ignition system on modern cars is part of the emission controls, and is subject to the strict warrantee period in California and some other states.

The fact that 20 year old ECUs fail is no surprise to me. They were built with an embedded design flaw(well known) that causes them to fail in a well defined mode.

The peak(not RMS) voltage of an unregulated auto alternator may be around 20 volts, after going through the regulator, even a damaged one with some burnt diodes in the bridge the voltage will not get up to anywhere near 20VDC. That's why the 21 volt spec for short duration. This voltage was selected in part by the physical properties of the batteries of the day(mid-late 50s), and partly because of the breakdown voltage of human skin which is quite variable but is almost always above 25VDC. Now, some aircraft have 28VDC systems, but never in cars.

It's sad that you compare high current numbers that sound like lethal VOLTAGE numbers, and use an exclaimation point too. I can, and have held leads of over 100 amps at less than 12VDC safely(don't try this at home) because the breakdown voltage of human skin is high enough to impede flow.

A failed battery is a failed battery. I think we need to make a few general assumptions here, that the battery is in moderately good condition. If you have an open in the battery, first, you can't start the car with that, as there will be no voltage apparent at the terminals. If you have a shorted cell, or plates touching, the battery will be weak, but the short will actually increase the spike damping, not decrease it, because it will present a larger continuious load to the charging system.

I am not in the selling of ECU or suppressors business. Just an 'ole EE trying to keep the uninformed from making mistakes with their money, like buying 'gas economy pills' to improve mileage.

Ask yourself this; When was the last time you heard of a car radio, amp, CB, radar detector, EFI system(other than the LH) or other electrical accessory failing from a known overvoltage?

Doc

 

Claim: Protects battery, charging system, engine computer and 12 volt accessories when jump starting.

http://www.harborfreight.com/cpi/cta...emnumber=91478

 

All I know is that when my '83s regulator failed the volt gage pegged out in the red, the dash and headlight went really bright and then after a couple min, one of the headlights burnt out. This happened at least 4 times before I pulled the alternator and had it fixed. So I'm not so sure that it will only go to 15.5V casue the gage goes up to something like 17-18V and the gage was pegged so who knows how high it was. It was high enough to burn out a headlight, to me that's pretty high.

 

The main troulble with most surge solutions in a car is that sometimes you would rather burn out the headlights than shut off the power, even isolating the alternator.

Several sites talk about issues related to long battery runs and surge related problems from devices shutting off rapidly like fans or even high power stereo systems.

Its off my radar for now.

 

"The fact that 20 year old ECUs fail is no surprise to me. They were built with an embedded design flaw(well known) that causes them to fail in a well defined mode."

Totally false!!!!!!!

A perfect example is the old ('79) 928 analog L-Jetronic AFM type of ECM.
These units rarely if ever fail and these types were used in '78-'81 BMWs.
Many 911SC Lambda ECUs ('81-'83) rarely fail.

"It's sad that you compare high current numbers that sound like lethal VOLTAGE numbers, and use an exclaimation point too. I can, and have held leads of over 100 amps at less than 12VDC safely(don't try this at home) because the breakdown voltage of human skin is high enough to impede flow."

"because the breakdown voltage of human skin is high enough to impede flow."

What???????? Who said anything about shocks? It's an unregulated alt. which
is capable of over 100 amps that's the problem, few if any electronics survive.

 

Really? Gee it's on your own website:

Model 928 (LH-Jetronic) - 0 280 002 504 (506/7/8/9)
no-start, constant injector pulse, no fuel pump

Model -501 and 2 might be quite reliable, the same cannot be said for 504/6/7/8/9. Others in the 928 community know of it well also. It's a shame your so ill informed being in the 'business' and all.

On the other hand your comments cement my earlier assertion that voltage changes aren't a factor in auto electronics failures. You can't seem to win for loosing here........

 

"On the other hand your comments cement my earlier assertion that voltage changes aren't a factor in auto electronics failures"

Another ridiculous off-the-wall statement!!!!!!

Electronics DON"T have an inherent failure mode when supplied by a proper voltage.

"The surge allowance for auto electronics as mentioned is a DOT spec."

And it's only the real naive that "count" on that and have no idea about
what really happens under the hood.

Bottom line: Except for the 928 LH unit, the 928 electronics are very reliable
when having the proper alt./suppy voltage.

The next off-the-wall comments, please!

 

"Do you like gladiator movies Timmy?"

 

'79 is CIS. L-Jet starts in '80.

Sidebar: if you have an '80 - you can use '81-'84 brains. The opposite may not be true, for example, the mileage meter may not work. Wiring is the same '80-'84.