Alternator repeat failure

It does sound plausible. It will certainly be some kind of spike causing diodes to fail. This can also be caused by a bad connection in the alternator to battery circuit (positive or ground). The battery acts as a capacitor, dampening spikes from the ignition and other inductive systems in the car (to a certain degree).

If there is a resistance in the alternator to battery loop, the spike dampening function of the battery will go down. This is why most manuals and engine bay stickers tell you not to disconnect the battery with the engine running. I once fried an alternator this way, I never tightened battery terminals down, so it was easier to disconnect it again when a car was being put back into storage. On one of my cars the bolt tightening the clamp also connected the board system and alternator to the battery - not tightening that left room for voltage spikes. The alternator eventually turned into a huge current draw, draining the battery in a few hours.

Your shop can probably measure with an oscilloscope whether spikes indeed occur and what their cause is. Without that, it's guessing. Which may work - but may also cost you another alternator down the road. Once the cause is clear the cause can be targeted - with better wiring, a flyback diode in a smart spot (close to the source but behind a fuse), a capacitor directly on the alternator, an inline ferrite... they all can provide protection.

 

Hello Bill,
Agree completely with the advice given by 166&458 above. Loose battery cables could certainly do it. The A/C clutch could in theory do it (although it is a bit of a stretch), and it would not hurt to put a flyback diode across the coil. Good design practice calls for such a diode across every inductive load operated on DC. The higher the current draw, the more necessary the flyback diode.

I can think of three other causes for the repeat failure. One is - and I hate to mention the subject of endless discussions on this forum - jump starting. I assume that you have not jump started your Dino lately? The same voltage surges which sometimes kill the Dinoplex during jump starts, could also kill alternator diodes.

The second would be simply a manufacturing defect of the diode in the rebuilt alternator. Such defects almost always show up as failure within the first 100 hours of operation. Was it less than 100 hours, you think? If that was the cause of the failure, just curse your bad luck and put in a new one (easy for me to say!). Statistically pretty unlikely, though.

The third possibility applies to all Dinos, not just your car. Alternator failures have been more frequent in Dinos than in other similar cars. Why? The following is entirely my own speculation, so take it for what it is worth, but my reasoning goes like this: Similar or identical alternators were used by many other cars. A/C clutches without flyback diodes were used by many other cars. Brutal jump starting is done to many other cars. So why should there be so many alternator failures in the Dino, and only the Dino? I believe it is temperatures. Excessive temperatures can kill diodes. Only in the Dino is the alternator tucked away between the engine block and fire wall with almost no air circulation. Not only that, but the red-hot exhaust headers are practically wrapped around the poor alternator! Imagine the Dino idling in city traffic on a hot day with the A/C running. The operating temperature of the alternator housing must be completely off the scale! Last time your alternator failed, it was apparently during the summer. And now...summer. If you believe in this theory, try to improve the heat shielding around the alternator. Ceramic coated headers should also help to keep temperatures down.

Lars

 

Some of the racing Lancia Stratos put the diode packs outside the alternator for easy replacement.

 

this is a job for Jim Selevan! He will know the answer!!

 

Not sure how you mean in the wire, etc. but I think you have it right. I have tried to find John´s post #050911 that you refer to but I can not find any such number. There is another post which covers this issue. It is # 071103. Perhaps that was the one you wanted to refer to?

The flyback diode should be connected across the clutch coil as close to the coil terminals as possible. The white band should be connected to the plus side of the coil, the other side of the diode to the ground terminal of the coil.

 

I had a customer with a Dino spider who asked us to remove the entire AC system after he blew an alternator.

He lived in Southern Cal and never used it. He tried the AC once and the alternator blew immediately.

So, it is possible. I've seen it happen.

 

While all of the suggestions and observations are excellent, I am not a believer in jump-starting as a cause of failure. Furthermore, millions of vehicles of this same era where designed with electromagnetic clutches on fans and A/C compressors. The number of failures is relatively small compared to the potential universe of susceptible alternators.

This is not to suggest that I have a better answer. I have had alternator failures, but usually related to oil on the brushes, or heat from the headers.

The high-current diodes are typically rated for several hundred volts reverse bias. It just seems unlikely that battery jumping, or A/C clutch disengagement (inductive load and open circuit voltage spike) would lead to diode failure.

Is the regulator functioning correctly? While the test device may have offered "..bad diode.." as a culprit, I would first check the field voltage to see if it is inversely related to the battery voltage.

No magic answer. Tuff stuff, these alternators.

Jim S.

 

Bill,

I have read the John Corbani post and I understand what you were saying now. Yes, you are right. John put the diode close to the switch, since the starter solenoid is not very accessible. The diode will do the job close to the switch too, but placing it closer to the source is always better.

My only minor disagreement with John´s post is regarding the choice of diode. 1N4004 is tiny little device, more suited to electronics than big electro-mechanical devices such as the A/C clutch. You don´t want to take chances and have the diode short out. A 1N5408 would be ideal. It can withstand peak reverse voltages of 1000V and up to 200A surge current. It is one tough little diode. I don´t think Radio Shack normally stock it. They list 1N5402, which is the same device but with only 200V peak reverse voltage. If you know of an independent electronics store in your area, they usually have the 1N5408 (they are industry standard numbers). Otherwise Digi-Key has them:
http://parts.digikey.com/1/parts/412536-diode-std-rec-3a-1000v-do201ad-1n5408-tp.html
It will set you back a solid 38 cents, but it may be worth the investment.

Good advice from Jim, as always. It doesn´t hurt to verify that the tester is telling the truth. Most likely, though, you will find that it is the alternator diodes after all.

As for the cause of the repeat failures, excessive temperatures still seems to be more likely, but one should not rule out the A/C clutch either, since several users have reported that there seem to be some connections between the two. Trying to think of possible scenarios with the A/C clutch, I can think of two, which would make life difficult for the alternator: The thermostatic switch which controls the clutch could have a slow changeover instead of a positive snap action. That could set off a string of sparks (high voltage spikes) when it opens or closes. The other would be a few shorted windings in the clutch itself, causing excessive current draw and higher-than-normal voltage spikes. Measure the resistance of the coil, or the current going through it, and compare it to the reference (in the manual?), or the markings on the coil.

And by all means, install the diode. In either of those two scenarios, a flywheel diode (flyback diode) would almost completely eliminate the resulting voltage spikes. If it was voltage spikes from the A/C clutch that killed your alternator, the diode would have saved it. It would have been the best 38 cents you ever spent. If it was heat, well, that is another story.

Lars

 

Could it be the Voltage Regulator? I wonder if switching to an alternator with an integral solid state voltage regulator would help...
Mine has been great!
http://www.ferrarichat.com/forum/showthread.php?p=140747641

 

Good to hear that the alternator itself and the diodes are OK after all.

I think you are right that the capacitor is there for noise suppression in the radio, but it would also do a pretty good job of supressing the voltage spikes you talked about. This capacitor does not exist in my car - or at least I have not seen it - even though it has both radio and A/C.

If you still wanted to install the diode, you could crimp or solder the cathode (lead with the white band) to the green wire terminal, and let the diode lay along the green wire. Then solder a ground wire to the anode lead and connect it to a chassis/engine ground point. Put a piece of shrink tubing over the green wire/diode assembly.

Thanks for posting the picture of the alternator removal. I always wondered how it could come out that way. With great difficulty, the answer seems to be! It does not look like much fun.
I admire your tenacity. Keep us posted.

Lars

 

Wow Bill they really do put the alternator in a silly spot!

 

Yes, it is a cheap precaution. As an added benefit, the switches will also last longer, due to less arcing.
It is probably not worth doing with the various motors, though. The simple diode approach does not work very well with motors. There you would need to switch in a resistor to dissipate the energy and it would be too complicated to retrofit. It is the solenoids and clutches which are the real offenders. If you have diodes across both the starter solenoid and the A/C clutch you should be in good shape - better than most other Dinos!

Lars