Strange alternator(?) failure. Need ideas

[cstreit]

Gang,

We were at the track this weekend with the racecar. Car ran fine through practice and qualifying. 3 laps into the rainy race the battery light comes on. (LOW VOLTAGE) 83 SC engine "base" with Paris Rhone 75AMP internal regulated alternator.

I assume the worst (broken fan belt) and immediately pull in. Turns out the belt was fine. The field wire (blue one from alternator) MAY have been disconnected. I have a quick disconect on it for fast maintenance. Well I reconnect it. Start the car, and no dice. Still no charging. Stumped.

So I need some help as I'm worried that because I don't know what the problem is, replacing some parts may not fix it. Here was my diagnosis process:

NOTE: I do not use the standard alternator lamp (no gauges). I use a small resistor to bootstrap the alternator. It has always been "hard starting" requiring me to bump the engine above 3k RPM before it will start. After that works fine all the way down to idle.

1. Light comes on after 3 laps. Pull in. Check and possibly reconnect field wire.
2. Light still on.
3. Recheck connections. Positive, ground, field wire all good.
4. Check ohms on field wire up to the front of the car. Shows expected 100 Ohms
5. Check continuity of all wires, ground, connection to battery, etc.. All good.
6. Restart car, no charge.
7. Remove alternator. Find small rivet near front of housing (it was a pop rivet head that fell into fan). Remove it. Disassemble alternator fully to clean, look for damage. See none.
8. Put it all back in. ALternator starts working again so we assume it was the rivent shorting something.
9. Next day alternator dies after 3 laps again.


So I'm stumped. Normally when regulator goes bad you see high voltage. In this case I see NO charging. Once the car cooled down alternator worked for a bit.

Can you think of ANY other reason this would not be fixed by just replacing the alternator? I can't test/replicate in the shop and don't want to invest in a trip to the track just to find out I didn't fix the problem.

Thanks!!!

[ischmitz]

Could the brushes be nearing the end of their life? When they get too short the lack of spring tension makes them loose contact to the rotor. Also, keep in mind the whole assembly is much warmer when the engine runs on the track.

[cstreit]

Yes I looked a the regulator and the brushes were still pretty long.

I am also assuming the failure is heat related. More than likely disassembling it only "fixed" it temporarily as the time allowed it to cool down. Just don't want to spend another weekend and $1k on a race to diagnose the problem a second time you know?

[ischmitz]

Understood. If that is the case (heat) may there is either a cold solder joint somewhere at the diode pack or even some bad diodes. The thing is I can see that one diode can have a temperature-related failure but to make the alternator stop charging all diodes supplying the D+ terminal must be bad or all diodes for the main connection. That is unlikely.

With one diode of the main diode pack bad you would get a slight AC component and a glowing charge light but the alternator would still charge. I had that on my stock '74 Paris Rohne. In your case digg around the D+ connection.

[cstreit]

I don't have a charge light though.... WHen it was working I got 14.3 volts NP. WIll check that connection though... Also going to put a load on it to see what that does...

Also wondering if the field wire getting disconnected while running may have damaged the diodes?

[jpnovak]

Chris,

Have you checked continuity across your blue field wire? what power rating is your ballast resistor that replaces the dash "lamp"? You may want to replace the resistor as a matter of maintenance.


I agree with Ingo that you may have a temperature related failure. I would check the resistor and also the diodes.

Is it time to rebuild the alternator?

[cstreit]

I did check the field wire continuity per above. From the alternator to ground I got 100Ohms... The rating was for 12 or 20 volts at like 100Ohms I believe... I'd have to dig around and look.

Alternator has been on the racecar for at least 7 years seeing obviously hard life. I have no problems replacing it, just want to be sure that is the actual issue.

From my reading just disconnecting the field wire while running shouldn't hurt anything. Another weird one. I remember at one point after we reassembled the alternator I forgot to reconnect the field wire, and the alternator still started.... (at 3k RPM)

[cstreit]

Wondering here... The resistor I put in was fairly small, leasing me to believe maybe the Wattage was low. Can't find any 68ohm 4 watts, but did find this 68 ohm 5 watt version. Would this be acceptable?

Amazon.com: 68 Ohm 5W Resistor Wire Wound 5% Tolerance: Electronics

[jpnovak]

Chris,

The alternator should require 60-100 mA of current through the field wire to excite the windings and generate current (Thanks Warren for the reference in archives).

by my calculations your 100Ohm resistor would generate 140mA of current. This should be more than enough to get things going. However, the 140mA generates more than 2W of power. This would cause the resistor to over heat.

Where is the resistor located in the car? Is it on the dash near your indicator or at the alternator. The heat and vibration may be causing it to fail. You could pull it out and heat it with a hairdryer while measuring the resistance to see if it fails under a thermal load. This is fairly common.

You might also be experiencing a failure in the internal regulator on the PR alternator. This has been known to happen before too.

[cstreit]

Good idea. I'll put the meter on it and the heat gun. it is the small type not the larger square ceramic version... It's up in the "boot" Wondering if it was too small....

[cstreit]

30 seconds with a lower watt heat gun only dropped the resistor from 98 to 94 ohms....

[jpnovak]

OK. Good results. I think the resistor is good.

Time to dig deeper in the alternator and voltage regulator.

Given your alternator output was steady at 14.3V then I would look more at the internal diodes. You can check diode behavior using the heat gun as well but you have to isolate each diode. Maybe just try to reflow the solder joint to each diode.

[304065]

Chris, now you know why I HATE internally regulated alternators

My Carrera alternator did the EXACT same thing at the Gingerman Club Race-- causing me to pit in during the enduro, thinking broken fan belt! Fan belt fine, ran the rest of the race on the batteries!

When the internal VR overheats it shuts down! It's probably 4" from the top of the engine case which is probably 250 degrees

Use an externally regulated 55A bosch and put the VR in the cockpit up by the switches!

[Grady Clay]

Chris,

Your 3W, 68Ω should be fine. (Worst case (14V) /68Ω = 2.9W)
Measure it. Heat it up. Wiggle the wires. Replace the bulb. Inspect the socket & wires.
Is the resistor glued down to some insulating base (little carbon resistors are prone to vibration damage)?
If you change to one of the large ceramic/wire resistors, glue it down and put a loop in each wire. (I like these much better.)
I would be concerned about temperature and age related issues. Use a heat gun.

Measuring the AC voltage will usually tell you if you have a failed diode (or broken solder joint).



I agree with your not liking $1K+/day alternator test sessions.

Swap alternators in a street 911 for testing.
Swap another alternator into the racer.
If you have a failure, be prepared to recharge the battery between sessions (parallel two big batteries for longer sessions).

I have seen frustrating situations where replacing the measures-good battery solved the problem (for unknown reasons).

Best,
Grady

How long are your races?
Have you considered doing without the alternator?
Without an alternator, it is lighter weight, nothing to fail, less rotating mass, less load on the fan belt and more.
There is a simple air pressure switch to warn of a broken belt (separating the broken belt function from anything else and allowing a very bright warning light).
Simply recharge the battery between sessions and use a ‘starter battery’ for cranking.
I use a little Honda generator to keep everything charged.
G.

[cgarr]

Subscribed!


Sent from my iPhone using Tapatalk

[Grady Clay]

Chris,

John is exactly correct, when using an alternator, use the smallest (low Amp version) that will handle the electric load.
Have a separate wiring harness (usually necessary when using a ‘cut-off’ switch.
Mount the external voltage regulator in the cockpit.
Don’t confuse the issue of a broken fan belt with alternator/VR failure.
kiss.

Best,
Grady

[cstreit]

Thanks for the insight guys.

I'm not really looking to diagnose the alternator per se, but diagnose that it's NOT something else that will lead to another track failure. The resistor for example. If it were failing all the other alternators in the world wouldn't fix the issue.

Other than the alternator, resistor or loose wiring, is there anything else that would cause this suddent failure? (Remember I don't have an alternator light.)

Grady, links to the air-pressure switch?

[Grady Clay]

Here is the 908 air switch:
I have it adjusted for the fan belt warning light to be on below about 3000 rpm.



Here is the manufacturer:

FRABA Posital GmbH
Schanzenstr. 35
Koln 51063
GERMANY

Member since: 8/15/97

Designated Representative:
Klaus Matzker
Phone: (49) 221 9621328
Fax: (49) 221 9621320

Data sheet:
http://www.posital.com/products/dw/DataSheet_DW_20040213_en.pdf


There is another cute little part:
An adjustable oil pressure warning light switch that is a direct replacement for the standard 911 version.
It turns the low oil pressure warning light on at pressures between about 10 psi and 60 psi.
I keep mine adjusted about 45 psi.
With the oil up to temperature, the light comes on below about 4500 rpm – perfect for an engine that never runs below about 6500 rpm.
Porsche Part number 771.613.572.00.

Best,
Grady