Pelican Parts
Parts Catalog Accessories Catalog How To Articles Tech Forums
Call Pelican Parts at 888-280-7799
Shopping Cart Cart | Project List | Order Status | Help

Go Back   Pelican Parts Forums > Porsche Forums > Porsche 914 & 914-6 Technical Forum

LinkBack Thread Tools Rate Thread
Thread Post New Thread    Reply
Registered User
pbanders's Avatar
Join Date: Jan 2001
Location: Phoenix, AZ
Posts: 1,113
Send a message via Yahoo to pbanders
One more forward step in my idle obsession (very, very long)

Sorry this is soooo long, but the background on this is important...

Most of you know that I've been obsessed with getting my 2.0L to idle properly. What I mean by "properly" is that once the car is warmed up, the engine speed returns rapidly to idle when the throttle is closed, that it doesn't "bog", and that it stays stable at a setpoint of about 1000 rpm, with minimal dipping (50 rpm or less) due to electrical load.

For a long time, my idle has been strongly affected by electrical load. If I turned on the lights, I'd get a significant (~200 rpm) drop in the idle. If I blipped the throttle, the engine speed would bog first before returning to the idle level. The engine didn't run smoothly at idle, I could feel slight misfires. I replaced the alternator, checked all of the grounds in the car, cleaned all of the body ground points, replaced the battery. I cut out the first 1/2" of my positive cable strap, exposed clean copper and reconnected it. I've got the copper strap ground in the car, checked it visually and electrically. I replaced the ignition with a Crane, went through my D-Jet, got all new components. Tested the ground and positive battery connections for proper voltage, current, and resistance levels. Replaced my voltage regulator with new OEM, then later, with a solid-state model. Cleaned all my fuse connectors, checked every positive and ground connection I could find in the ignition, starting, and FI system. Bought and built a Heathkit CO meter to set my CO, played with a zillion different idle mixtures and bleed screw settings. Checked every spot I could for vacuum leaks. Analyzed the PCV system to see if it was causing the problem, tried different ways to connect the system into the plenum and air box. Through all of these changes, the idle would get better from time to time, but the problem never really went away.

I got my engine rebuilt recently, and figured that would finally fix it. At first, it seemed to - but after a couple of weeks, the problem came back. Now, in addition to bogging, my idle would "float" - it would hang at about 1300 rpm until I dragged it down to 1000 rpm, by slightly releasing the clutch while in gear. I retested everything, didn't see anything that would explain it.

My car had sat for about a year in my garage before I took it in for the rebuild. I had bought a new Diehard battery that sat also for about a year. I had a cheap Schumacher manual charger, so before I took it in for the rebuild, I charged it at 10 amps for a few hours and it started up. When the shop finished the rebuild, the charged it for a bit, and stuck it in, and it started up.

One recent evening, when the problem started coming back, I was playing around with the settings for a while, trying to get it right. I shut it off for a few minutes and came in, came out to start it to pull it into the garage. It cranked once and that was it. Battery low. Hmm. I'd just driven it to work that day, it should have gotten a good charge. I put it back on my charger for a few hours and the battery seemed fine.

Bad battery, right? Hold on...

Finally, a couple of nights ago, I got in some of those new fiberglas speaker boxes that are sold on Ebay for the 914. They let you put a full-sized round 6 1/2" speaker into the speaker box position. I've already got 6 1/2" speakers in the doors, thought this might provide equivalent bass to those and pump up the system. Well, they look nice, sound fine, but compared to my door speakers, the box volume is just too small for decent bass. The combo of both, however, is very nice. I ran the system on and off for about 20 minutes with the car off, and planned on driving the car to work the next day.

I went out to start it the next morning - and you guessed it, the battery was nearly flat. Cr@p. I drove my M3 to work and decided at lunch to get a decent charger that would tell me when the battery was fully charged. I bought an automatic 2/15 amp Shumacher charger at Autozone at lunch, drove home, put the car on the charger, and went back to work after I had lunch. I noticed as I left after about an hour that the charger was still at 15 amps. BTW, I removed both battery terminal straps before charging the battery in-car.

I called my wife a few hours later to tell her to make sure the kids didn't disturb the setup as the got their bikes out. I had her look at the charging level - still high, at 12 amps. Wow, that battery was really discharged.

When I got home, the green "fully charged" light was on and the charger had switched off. Cool, now let's hook that thing back up....

The copper ground strap SNAPPED OFF where it attaches to the clamp the moment I moved it towards the terminal.

In that moment, it all was clear what had been going on with my car for years. The battery ground strap was likely bad for ages. Testing didn't reveal the problem, nor did the visual. It had to fail completely before I found out. Add in the fact that I've probably never had a fully charged battery in the car in over a decade (due to my not using my old charger long enough to fully charge it), and it explains why my alternator load on the engine has always been high. And the problem is worse at idle, where less power is available and engine speed is lowest. The reason it ran well right after the recharge was that the shop charged the battery much more (but probably not fully) than I had. It took a couple of weeks of poor supply due to the crappy ground to finally let the battery droop to a level where my idle sucked again.

I went back to Autozone and bought a $6, 2 guage, 12" battery ground strap. Connected it up, started the car, and went out for a drive to bring the engine to a fully-warmed up level. First thing I find is that now, my idle is about 400 rpm too high. I screw in the bleed screw to bring it to 1K rpm. Stable as a rock, no missing at all. I now start adding electrical load. Headlights. Hi-beams. My H3 fog/driving lights. Heater blower. Fresh air blower on high. Stereo on high. Total idle speed drop was about 50 rpm. Blip the throttle. Comes down smoothly, doesn't bog.

What is there to learn from this story? The very first time I started debugging this problem, 10+ years ago (no kidding), I thought the problem was due to drag on the motor from the alternator. My voltage levels always seemed ok, so I went after the charging and battery system. When I didn't find the problem, I quit looking at the charging/battery system and went on a wild-goose chase through my ignition and FI - even though the symptom was still staring me in the face as being in the charging/battery system.

Moral of the story is to be methodical and if logic is telling you that what you're seeing doesn't make sense, then keep at it until you figure out why, instead of going off-track. All that stuff I did on my ignition and FI system to fix the idle didn't do a thing to fix it - though it sure runs nice now!

My next step is that I'm going to replace the tranny ground strap with a brand new strap as soon as possible. If you've still got the 35 year-old original battery and tranny ground straps in your car, no matter how good they look, I strongly suggest you replace them. And if you have never made sure your battery was fully charged, taking it to Sears and spending $20 to have them analyze and charge it might be a good idea, too.

Last edited by pbanders; 01-05-2007 at 08:20 AM..
Old 01-05-2007, 08:17 AM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #1 (permalink)
Registered User
Ray Mital's Avatar
Join Date: Apr 2006
Location: Park Ridge, NJ
Posts: 243
I like a story that has a happy ending....thanks for sharing.
Looking forward to checking my grounds.
'73 914 2.0L driver
'73 914 2.0L parts
'74 914 2.0L project
Old 01-05-2007, 12:25 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #2 (permalink)
Banned for Life
RSBob's Avatar
Join Date: Jan 2003
Location: "Apple Maggot Quarantine Area', WA.
Posts: 5,488
Important lesson for the rest of us. Watch that your ground strap is good and securely connected.
Rick S, '71 914 3.0, '82 911SC EFI, R Gruppe #309
Old 01-05-2007, 08:30 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #3 (permalink)
Registered User
pbanders's Avatar
Join Date: Jan 2001
Location: Phoenix, AZ
Posts: 1,113
Send a message via Yahoo to pbanders
Like Mike Zois said to me, don't declare victory over any problem with your 914, you'll just jinx it....

OK, so after charging my battery fully and fixing my battery ground strap, I finally had a stable idle - incredibly stable - under any electrical load I could throw at it.

So, after it was all done, I ran the car for a while, then shut it off, and about 30 minutes later, measured the battery voltage. I'd started at 13.1 V right after the charge. Now, it was 12.5 V. Ran it some more on the street (not just idling), measured again, 12.4 V. Measured it while the engine was running - 12.4 V. Something was wrong, it should have been over 13 V if the alternator was charging. BTW, my idle was still super-stable with electrical load at this point.

Did a series of tests on the battery after recharging it to full - nothing wrong with the battery. Tried three different voltage regulators (2 mechanical, one solid-state) - same situation, no charging voltage. Measured the D+, DF, and D- pins of the alternator socket at the relay board - and that's where I noticed something odd.

The D+ pin was as I suspected, low and at the same level as the battery. DF at idle, was showing almost no voltage - not surprising, the battery wasn't charging, no field was being generated. But D- was odd - it was at - 1.1 V. If you look at the circuit diagram for the relay board:

... you'll see that D- is at ground - see part 4? That's the voltage regulator, on the bottom of the diagram, you see the ground symbol that's tied to the case of the voltage regulator, and you can see that D- is tied to it.

There's just one problem - WHERE IS IT GETTING THE GROUND? The case of the voltage regulator is attached to the side of the relay board with two screws - but the relay board is made of PLASTIC. It's not tied to ground. On other VW applications, like the Bus, the voltage regulator is connected to metal (which also helps conduct heat away).

So, I didn't immediately pursue this, instead, I put my oscilloscope on the D+ lead to see if I had a bad diode. I saw what looked like a problem with a bad positive diode. See:

This is a great reference (if you interpret it right - see below). My output looked like the third scope tracing, "short circuit of positive diode". Looks like I need a new alternator. So, I cleaned up and planned on getting a new alternator on Monday.

As I was getting ready for bed, I thought more about that D- being at - 1.1 V. That can't be right. There must be a significant resistance to ground. I went out and quickly tested the resistance from the case of the voltage regulator to chassis ground - it was 4 ohms, way too high.

I went back inside, it was getting really late. I wondered - what if I provided a ground wire to the voltage regulator case? I got up this morning and started experimenting. First, I measured the resistance again - it was 4 ohms. Then, I pulled the alternator plug - the resistance went up to about 100 ohms. I could see what was happening. The voltage regulator D- current path to ground was mostly through the alternator. If I pulled the alternator plug, now the path to ground was back through the voltage regulator to D+ and up through pin 2 on the relay plate, which is the alternator warning lamp circuit (used to provide the initial current to start the alternator) - which has a resistance of about 100 ohms. It was as I suspected - there was NO case ground for the voltage regulator because the case was connected to a plastic relay board plate.

I then made a short ground jumper wire from the screw that holds the relay plate to the chassis and to the screw that holds the voltage regulator to the board. Measured the resistance from the voltage regulator case to ground again - now, it was 0.1 ohms or less. Now, for the big test, I started the car.


At idle, D+ was now at 13.5V, DF was about 5 to 6 volts (because the alternator was working and supplying current), and D- was at under 0.1 V. My alternator was working, my battery was charging. There was nothing wrong with my battery, voltage regulator, or alternator - the problem appears to be in the design of the system by Porsche. There's no ground for the voltage regulator case, which pulls down the charging voltage by about 1.1 volts. This lets the battery drop down to about 12.0 V or so before it starts to get charged by the alternator. By providing a hard ground to the voltage regulator case, DF is clamped to 0 V and the voltage regulator works properly.

So, that's all great! My battery is charging and I don't need a new alternator! Except for one thing - now, my idle is definitely affected by adding electrical load to the system!! ARRGGHHH!! Now I understand what was happening. When I charged the battery fully, I brought it well above the 12.2 V to 12.0 level where the charging system would start to charge the battery, so I had no load on the alternator (as evidenced by the near 0 V levels I was reading on DF before I discovered the ground problem). All of the electrical load I was adding was being sourced by the battery - no wonder the idle was so rock-stable. Now, by fixing the charging system, it was supplying voltage to DF at idle, to get a charging voltage of 13 V or so. This caused a load on the alternator, and pulled down my idle.

I made sure the car was fully warmed up, and got out my Heathkit air/fuel meter. I was running rich at about 4.5% at idle, so I dropped it down to 3.0% - right where the mark on the ECU surround is notched by Bosch. I'd found in the past that when I had the mixture rich, it made the idle droop under electrical load more sensitive. I tweeked for a while, and got it to a pretty acceptable level of droop, about 100 rpm or so - I could live with it.

Sometimes, I wonder if these cars are possesed!! So, if Mike Zois is reading this, I'll say it:


I don't want to jinx it.

Last edited by pbanders; 01-08-2007 at 07:40 PM..
Old 01-07-2007, 08:24 AM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #4 (permalink)
Registered User
pbanders's Avatar
Join Date: Jan 2001
Location: Phoenix, AZ
Posts: 1,113
Send a message via Yahoo to pbanders
Well, with the help of others, the problem on my car isn't a design flaw with the charging system - I have a bad alternator ground. Despite what the wiring diagrams show, in a properly set up 914 charging system, the VR case is NOT grounded by the relay board, and D- in the alternator is tied to a case ground. In my car, the alternator case ground is bad, and by tying the VR case to ground, I bypassed the fault. I just had my motor rebuilt and my mechanic did the alternator harness, so I'm going to start looking there, as well as at the tranny ground strap.
Old 01-08-2007, 07:06 AM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #5 (permalink)
Dave at Pelican Parts's Avatar
Join Date: Dec 1969
Location: Silly-Con Valley
Posts: 14,061
Send a message via AIM to Dave at Pelican Parts Send a message via Yahoo to Dave at Pelican Parts
Don't forget the ground path through the case of the alternator to the bracket, from the bracket to the fan shroud, and so on. There was lots of corrosion between my bracket and fan shroud at one point, and it gave me charging problems.

Pelican Parts 914 Tech Support

A few pics of my car:
Old 01-08-2007, 03:36 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #6 (permalink)
Registered User
pbanders's Avatar
Join Date: Jan 2001
Location: Phoenix, AZ
Posts: 1,113
Send a message via Yahoo to pbanders

If you have an alternator ground fault, and you add a ground from the VR case to ground, ALL OF THE CURRENT GENERATED BY THE ALTERNATOR WILL FLOW THROUGH THE JUMPER WIRE, 20A or more. You could start a fire. DON'T ADD A GROUND JUMPER.

Instead, if you suspect you have a problem, measure the voltage at D- while the car is running. If it's more negative than -0.5 V, then you have a bad alternator ground that must be repaired.

Bosch/Porsche didn't ground the VR case for a reason. They must have recognized the potential for a problem if the alternator ground was faulted, and decided to let the VR case float instead of providing an alternate ground path that can't handle the current level. I should have known - not only was it not a design flaw, it was a safety feature!
Old 01-08-2007, 07:45 PM
  Pelican Parts Catalog | Tech Articles | Promos & Specials    Reply With Quote #7 (permalink)

Thread Tools
Rate This Thread
Rate This Thread:


All times are GMT -8. The time now is 08:19 AM.

Powered by vBulletin® Version 3.8.7
Copyright ©2000 - 2019, vBulletin Solutions, Inc.
Search Engine Optimization by vBSEO 3.6.0
Copyright 2018 Pelican Parts, LLC - Posts may be archived for display on the Pelican Parts Website -    DMCA Registered Agent Contact Page

DTO Garage Plus vBulletin Plugins by Drive Thru Online, Inc.