3 fuse holder in engine bay

[NorthernThrux]

Posting this so that it might help some other people chasing down battery drain issues.

Ever since I got my car back, I have always left it on my CTEK charger and not had any issues with driving up to 4 hours with a few stops. But during the recent gorgeous weather, I put my Macan on the lift and have been driving the 911 every day. Well, on day 3 the battery was at 8V and while the lights and other things would come on, I didn't even try to start it before attempting to charge it.

Once charged and on maintenance mode on the CTEK (took 2 days), I parked it on the lift and then started monitoring the battery voltage. It was dropping by about 0.1 V every couple of hours (after bleeding off the surface charge using the light so the resting voltage was 12.75 V as a starting point). Almost brand new pair of Odyssey PC925. Unlikely to have failed. Hmmm.

So I opened the trunk and pulled all the fuses and measured the current across each pair of fuse terminals with my Fluke DVM. Nothing. No parasitic draws (with the trunk bulb removed). For interest, the trunk bulb draws about 0.5 Amps. The clock, under 1 mA (the minimum my DVM can measure).

With every fuse in the trunk pulled, voltage still dropped by 0.1 V every few hours. Hmmm. Disconnected the starter cable and measured the current draw. Few tens of mA. Odd. Failed diodes in the alternator? Bad starter motor solenoid? Both had been rebuilt, but you never know. Using a cigarette lighter voltmeter I only ever measured 13V at 3000 rpm, so a bad diode was a possibility (more on this voltage level later). But before pulling the alternator to test individual diodes, I tested the whole thing with my diode tester at the B+ terminal and ground. All good. I disconnected the lead from the starter solenoid where the alternator and the battery leads come together and bolted them together (i.e. removed the solenoid from the circuit). No difference. Battery still dropped by 0.1 V every few hours. Hmmm.

Then I pulled the fuses in the rear fuse holder (I have 2 in what I though were the 3 and 2 positions). Success, no parasitic draw. Battery lost 0.01 V over 30 hours. That's more like it. I measured the voltages at the two red wires feeding the fuse block. One was 12 V on always. One was switched 12 V. As they should be. Then I measured the current draw across the fuse terminals and there was a parasitic draw across the top fuse connections in the block (should be fuse 3) and nothing in the spot below. Well, that's odd. The #3 fuse should be going only to the 2-stage defrost relay and that isn't on when the car is off. Pulled the relay. Still had the draw. Hmmm.

Then I looked at the fuse block. It was inverted!!! #1 fuse on top and #3 on the bottom. If you look at the circuit diagram, you see that the switched line normally goes to the MFI rpm switch and the cold start relay. The unswitched line goes to the defrost relay. If you invert the fuse block, then because #1 and #2 are bridged internally in the fuse block, the unswitched voltage is now going to the cold start relay (not the relay solenoid, but terminal 30 that gets switched to the cold start valve when the starter is engaged on a cold engine) as well as the MFI rpm switch. The switched line went to the sporto blue/yellow line which does nothing and has no fuse.

So the upshot of this was that the battery was powering the MFI rpm switch all the time, and further, the cold start solenoid was being inconsistently switched due to the large voltage drop on that line when the starter was turning. I removed the console, flipped the fuse block and reinstalled everything properly.

No more draining battery. It lost 0.03 V over 5 days. And cold starts are instantaneous (well instantaneous for a 47 year old car, so about 2-3 s of cranking in reality) without the added distraction of having to floor the accelerator (which I had to do some times but not others - probably related to battery voltage).

Moral of the story. (1) Even an excellent restorer can goof up once in a while. This one is hard to catch since the car works more or less as it should, with only intermittent cold start issues. (2) And if you do have any significant battery drain, check the fuse block. It might be upside down or even if it's not, the switched and unswitched wires could be reversed. When wired as god and Dr. Porsche intended, a good battery should last a whole winter without charging at the discharge rates I'm seeing (although I will always leave it on the CTEK charger in AGM maintenance mode).

One last thing. A month or so ago, I made a post about installing a modern voltage regulator that looks like the original SEV Marchal tin can. https://www.early911sregistry.org/fo...=1#post1067854 . Turns out that was a dumb move. Because the VR senses the set point voltage before the two battery parallel isolation diodes on these alternators, it you install a 14V set point regulator (like the one sold everywhere for our cars and in that post), you only get 13.2 V at the battery or 13.1 V at the cigarette lighter at 3000 rpm (12.5 V at idle). Not really enough to charge modern AGM or lead acid batteries. I looked up the specs of what had been in there. It was a Transpo ID4010. It had a 15 V set point. Whoever put that in in the distant past knew what they were doing. This gives 13.7 V at the cigarette lighter at 3000 rpm, which diminishes to 13.1-13.2 V after driving around for 15 minutes, as the battery recharges after starting. You get 13 V at idle at the cigarette lighter, which is decent. I used the same trick as in that post and put the SEV Marchal can over the Transpo. You'd have to look long and hard to realize it wasn't the original VR.

Final product

Left side	Fuse position	Right side
Red unswitched (very faded red)	#3	Red/Blk (mine was so faded you could only see red/black if you dug into the harness). Goes to defrost relay.
-	#2	Red/Wht (to MFI and cold start relays)
Red switched	#1	Blue/Ylw (sporto valve - unused - no fuse on manual tranny)

You'll see an extra thinner red wire on the unswitched supply at position #3. That's my battery charger lead. That unswitched line goes via heavy gauge wire to the alternator/starter solenoid and directly on to the battery. Handy.

I do need to order a set of proper copper ceramic fuses for the car, but for the moment, it all works nicely.





This should keep the batteries charged in daily driving!!!!

[911kiwi]

Thanks for sharing, Ravi! That was very insightful. You are a mine of information - GOLD MINE, that is!

[davep]

Excellent detective work as usual. However you may be partially wrong on the clock current draw; does it not periodically power up to do a rewind, and then sit idle for long periods?

[Nicklasliljegren]

Ravi,

Pls can you post a photo of your voltage regulator

Best Nick

[NorthernThrux]

Excellent detective work as usual. However you may be partially wrong on the clock current draw; does it not periodically power up to do a rewind, and then sit idle for long periods?

Yes, dat true. That’s why when you remove the fuse, the clock runs for about a minute until that spring unwinds. If you then hook up an ammmeter it will catch the clock rewinding the spring. You can hear it. Still less than 1 mA, at least with whatever the response time of my Fluke DVM is. The rewind sound lasts about 2 seconds.

[G69]

Somewhere back there isn't there also a pigtail that accepts a remote start button for dealer service? I would love to have that.

On my '69, the bridge is visible on top of the fuse block and energizes three of four circuits. I was also confused with the orientation of the fuse block.

I have found that the original wiring diagram is a starting point only-
Now I know that is because it is drawn for the German market cars. Export cars have significant differences.

Yes- copper fuses are very important. The lead-like ones can be a problem.

Copper is a large component of the brass fuse tabs on the fuse block and lead fuses and brass don't get along when dampness is present.

The cheaper lead/plastic ones are far more subject to corrosion because of the different metals.

Copper fuses behave much better.



As important is the condition of the solder-tinned wire ends that are attached to the fuse block by the set-screws.

We have all seen burnt fuse holders and that is usually from poor connection resistance melting the base at the set-screw and tinned wire connection.

When possible clean and re-tin the ends if suspect. This is when a big solder gun like the old-school Weller and some real rosin core LEAD Solder comes in handy. The job doesn't go so well with modern 'safety' low lead solder.

It is necessary to tin the end of any new or re-stripped wire so that the set screw has a single conductor to mate to.

A bare wire will separate at the strands and resistance increases. The strands corrode faster because their surface area is more exposed and there is also capillary action that wicks moisture up the wire under the insulation making a new end too corroded to solder.

As the connection ages it is a real fire hazard.

[NorthernThrux]

Good advice on the wires. And I love those old school heavy black Weller guns. I still have my dad’s, which he bought in the early 60’s.

Sadly no pigtail back there, at least on a ‘73. There is an unused green/white and brown 2-pole connector, but that’s for the optional rear fog.

Nick, the car is on the lift over my Macn. I’ll post some pics of the VR next time I lower it.

Ravi

[Nicklasliljegren]

Ravi,
That would be very much appreciated. The voltage regulator on my 911 E does not work. It seems difficult to find a NOS or to find someone who can restore the old one (a SEV Motorola). So your solution is therefore very interesting.

Best Nick

[Simonjjb]

Ravi

Where did you buy the Transpo?

[NorthernThrux]

Nick and Simon, The Transpo ID4010 was on the car when I bought it. It's uncommon in North America, but can be found in Europe. Here are some sources for the Transpo or equivalent. Not surprising, there are mostly EU vendors. What defines it is the 15V set point. The diode drop of 0.7V across the protection diodes means that you should be getting something like 14.3 V at the battery. Of course these things are all nominal, and I see 13.8 at the battery. Close enough.
The actual form factor is identical to the post in my build thread referenced above.

https://canadianautopart.com/catalog...ategory/14053/
https://www.energyparts-shop.com/en/...transpo-id4010
https://www.deer-online.com/alternat...-en_id4010_trs

This is a useful reference for the reason the Marchal/Motorola and the Bosch require different voltages. The Bosch one is the one shown in the usual wiring diagrams. It has 3 sense diodes that feed the D+ (effectively a low current draw tap of B+) signal to the VR. The Marchal produces the B+ from the actual battery feed (which makes more sense from tracking failures because if you lose voltage and have AC ripple, it is a direct indication of the failure of the main rectifying diodes. For the Bosch it could just indicate a failure of the sensing diodes). But for the SEV there are two diodes in parallel past this point (D2 in the diagram which is actually 2 parallel diodes that are on that red plate on the Marchal alternator at the back which produce an addition voltage drop of 0.7 V. So you need the higher set point of the proper 15V VR to get the proper battery charging voltage and not something 0.7 V too low.