Some of you know I recently converted my car to all LED's, to reduce the parasitic drag on the engine from the alternator at idle, when the lights and other accessories are turned on. I did some measurements with the car off, went out and got it fully warmed up (it's 95 deg. F here today, didn't take that long), then I took a series of measurements of the current supplied to the battery (from alternator B+) and the supply to the car from the battery, using my digital loop ammeter. Did some calculations, and here are the results. Battery was fully charged before staring this test (12.75 V with the engine off), idle steady at 1000 rpm, and battery voltage at idle was 13.7 to 13.8 V. Each accessory was turned on individually and the current for the accessory was calculated by subtracting the total current with the accessory off at idle. Current draw for various combinations of accessories was calculated by adding the accessory currents to the total current with all accessories off at idle. Note the table below is the current supplied to the lights and accessories (as well as the ECU, AAR, Fuel Pump, and ignition), and NOT the current coming from the alternator B+, I'll get to that later. You get the idea, if you have any questions, just ask. I'll explain the abbreviations in the table in a minute...

Click to view attachment

OK, here are what the abbreviations mean.

Rows
------
Lights (lo) = lights on, low beam
FA(3) = Fresh air blower, maximum setting
Heat = Heater blower
Wipers(1) = Wipers on 1st speed setting, flipped up, not dragging on windshield
Radio(20) = My radio on volume setting 20, quite loud, unit max is 22W x 4

Columns
----------
w/LED's = my current setup, all LED lights, including headlights and driving lights
w/LED's + hi + dl = same as w/LED's column, plus current for LED hi beam and DL's
w/Inc (lo) = as if I hadn't done the LED's, w/standard H4 halogen low beams
w/Inc (hi) + dl = as if I hadn't done the LED's, w/H4 halogen hi beams + H3 DL's

FYI, my old H4 halogen bulbs were 55/55W

Hope that makes sense.

First thing to note is the current with all the lights and accessories off. This is the basic current requirement for the car to run, supplying the ECU, the AAR heater, the Fuel Pump, and the ignition (I have a Crane XR3000 with one of their hot coils). This current is quite low, only 5.0 A. I also measured the B+ charging current from the alternator under these conditions, it was varying between 5 and 7 V as the voltage regulator was modulating the field current in the alternator, to keep the voltage in the system stable. I might note that I have one of the transistorized Bosch voltage regulators, instead of the electromechanical stock Bosch voltage regulator.

As you can see, even when I'm running full LED's and am just on low beam, if I turn on all the accessories, I'm pulling a lot of current at idle, as much as 26 A. However, this is still reasonably OK for the alternator, as it's spinning at 2X idle (2000 rpm) and at 80% duty cycle can put out about this much (I'll post some alternator data to this topic later about it's output at speed).

Try the same thing with H4 halogen low beams and you would pull as much as 43 A. According to Porsche, the max output of the alternator is 50 A, but I think Bosch says 55 A. Regardless, 43 A at an alternator speed of 2000 would be insufficient to charge the battery and would run the alternator at 100% duty cycle, not advisable. If you were dumb enough to be running your H4 halogen high beams and H3 fog/driving lights while at idle with everything on, that 59 A draw would cook your alternator in no time!

Prior to doing the LED conversion, what I saw was that the idle began to be affected when I just turned on the H4 halogen low beams (55W). As you can see, this corresponds to a current of about 25A, which with my LED's is only approached when I turn everything on, so the reduction of current by going to LED's lets me use my lights and accessories without any noticeable idle drop, for virtually all combos. And, that's what I've seen in actual driving, too.

Note that even if you weren't at idle, and you had on all the accessories, if you had standard lighting and H4's, the stock alternator at max output would barely keep ahead of the current draw, and you'd likely be pulling down the battery. Note that all of the currents in the table above are what's being provided to the lights and accessories, as well as the ECU, AAR, and ignition. I also measured the current coming from the alternator (the B+ lead), and as you might expect, it was about 2 to 3 A higher, because it was also charging the battery, so the actual requirements on the alternator are 2 to 3 A more. If your battery is in crummy shape or just flat from not being run, this could go as much as 10 A.

Anyway, I've never seen this data before so I wanted to collect it and understand what it means. I think you can see that for reasonable combinations of incandescent lights and accessories, that the stock alternator can be strained to keep up with requirements, and as a result, put a lot of drag on the motor as well as barely keep the battery charged. By going to all LED's you can keep the stock alternator and significantly reduce stress on it, making it last longer and improving the idle at the same time.

BTW, some obvious things you can figure out from the table are the currents for each individual component. I'll save you the trouble.

At 1000 rpm idle:

ECU + AAR + Fuel Pump + Igniton = 5.0 A
Parking, side marker, license, trunk, and instrument LED's = 0.8 A
LED low beams = 2.0 A
LED hi beams = 2.8 A
LED driving lights = 3.1 A
Fresh Air Blower, max setting = 7.9 A
Heater Blower = 6.0 A
Windshield Wipers, slow setting = 2.8 A
Radio, loud volume = 1.4 A

I had to do some calculations to figure out the incandescent draws, I used literature and ratings that I have for each of the components. As compared to my LED low beams, with all the other lights (parking, side marker, etc.) on, my old incandescent setup drew an additional 17. 1 A. The H4 high beams + DL's pull another 16 A!

That's some research there. I may yet modernize at least the rear lamps to
LED. I have my flame thrower Cibies out front though.
I wonder what the output of the 911 allternator is in comparison for the early sixes.
I'll have to check the specs on that. Anyway good stuff.
Thanks for the specs.

If you're going to continue to run incandescents, and you want to use high output H4's like the 80/130W bulbs, I suggest you go with the 160A alternator retrofit you've read about here in the forum. At least then you'll be sure your battery will stay charged!

Wow!
Thank you for the research and sharing.

Couple of other things this points out about old cars like our 914's and modern cars. Modern cars are engineered better, and also have very high electrical load demands at idle, due to the many additional systems a modern car may be equipped with. To handle these loads and keep the battery charged, modern cars use highly-efficient alternators with integrated voltage regulators, and the alternators are better sized to provide sufficient current under idle conditions. Additionally, modern engines are generally more efficient, and are able to handle load variations better than our old T4's with their somewhat antiquated design. But the most significant difference is the closed-loop idle control of modern engine management systems. When additional loading at idle is put on the engine, either from electrical loads or from mechanical loads like the air conditioning compressor, the closed-loop idle speed control provides additional air past the throttle plate to counteract the higher loading and keep the idle where it should be. When you're driving your modern Toyota, Porsche, or Ford, you hardly even notice the impact the constantly varying loading has on the engine under idle conditions.

It's been a back burner project of mine for years to incorporate active idle control into my 914's engine. I've got some ideas on how to do it, but have yet to put them to practice. It would be interesting to discuss various approaches to this and how it could be cheaply and simply implemented.

OK, so I fired up my Northwest Microfilm viewer (I think it was built in 1830) and stuck in my Porsche 914 Factory Workshop Manual fiche #6, "Electrical System", and went to page A-21 to look up the characteristics for the stock alternator.

The p/n for the 914 is 022 903 023.

Maxiumum Current = 50 A
Mean Regulator Voltage = 14 V
Rated Output Speed = 2200 rpm

Load Current Table
----------------------
Current RPM
10 1400
33 2200
50 6000

Note that all of the RPM here are twice that of the engine speed, because of the 2:1 pulley ratio. If you're running at an idle engine speed of 1000 rpm, the alternator is spinning at 2000 rpm.

Note that alternator output is NOT linear with rotational speed. There's a minimum kick-in speed where the alternator begins to work, then past that, output goes up rapidly and flattens out as you approach the maximum output at 6000 rpm (3000 rpm engine speed). That said, it's pretty clear that at typical idle speeds of 950 to 1000 rpm, output is right around 30 A or slightly less. Looking at my table of actual current requirements shows that for many combinations of lights and accessories, the car is operating very close to the rated output if not higher.

Brad, thanks for all of your work....I was just about to convert knowing it was a good idea but this really checks the boxes. Best, Grant

Will this info be added to your site as well?

Don't forget that lighter electrical loads also relate to better MPG (like anyone here cares)

I was contemplating a larger alternator, but looking at that chart, I am going to go LEDs.

Still would be interested in a more modern alternator with integral VR, if there was one that mounts easily. (No cutting as required with a bigger unit.)

Very interesting, thanks for sharing Brad!

...........and my 2000 BMW wagon has additional WATER coolling

Interesting question. I haven't edited my site in a long time, I originally used MS Frontpage to do it. I just need to find a WYSIWYG editor I like. As my wife is an expert web developer, I'm just going to see what she recommends. I'd like to update the site with a number of things, including this information.

I think the ideal (money no object, that is) solution is to go with LED's and also the 160 A alternator. Even better, if either I or someone else figures it out, add an active idle stabilizer, too. But it's clear the most cost-effective solution is to go LED's.

BTW, one of the next things I want to do is to duplicate some of the conditions in the chart I posted, while using an oscilloscope to monitor the B+ voltage, to see the degree of regulation taking place. From what I'm seeing with the B+ current variation on my loop meter, it's about 2 A of fluctuation, I want to see what the voltage variation is, and also make sure that all my alternator diodes are working correctly.

Brad,

This brings to mind an older thread as I am still contemplating the whole LED upgrade to my car as well.

If memory serves, didn't the stock configuration of the car prevent the high beam and fogs from coming on at the same time?

So wouldn't this change the theoretical max load to avoid the issue if you were still using all incandescent in the front?

I seem also to remember that the engineers knew this and if you had incandescent light or upgraded AND you did the J-west upgrade as I call it...you would also exceed the load of the alternator and possibly fry it.

This makes me think that if we wanted to upgrade to LED's one would want to do all of them...at least for the front of the car.

Thoughts?

Why?

Why do the mod, carry the weight and spin the weight if you don't need it?

I did some rewiring and added a couple of relays a few years ago to allow me to flash my fog/driving lights by pulling on the turn signal stalk when the lights are down and off, and also to allow me to turn on the fog/driving lights when I switch on the high beams. I've also got a small toggle switch I put on the fuse box to turn the simultaneous high + fog/driving light off, too. J-West came up with a similar mod the next year, I'd have bought that if it were available.

Basic automotive engineering tells you that at least with modern cars, you should design the charging/battery system such that when you're at idle, you can turn on every single accessory (at least all of the ones that run constantly), and not exceed the alternator's capability to supply the required current, while simultaneously providing sufficient current to charge the battery. Additionally, the active idle control should keep the idle stable at the setpoint, regardless of alternator loading and other mechanical loading (e.g. a/c compressor, P/S pump, etc.).

Unfortunately, that's not how the 914 was designed. From what I can tell, and I think my experience is shared with others, even with just the headlights on at idle, there's significant idle drop due to alternator loading, and because there's no active idle stabilization, the lower idle means the alternator is putting out even less current. Add in any additional loads (e.g. heater, fresh air blower, radio, etc.) and it gets progressively worse.

Going to LED's "fixes" this problem somewhat, but it's really a band-aid. The best solution is a combination of LED's, a higher capacity alternator, and active idle stabilization. Two of those three are available now, and I hope we can figure out a way to add idle stabilization in time to the 914.

Very nice, any calculations for voltage drop or increased current draw due to old wiring?

No, but it's probably on the order of 10 to 20 W throughout the whole system. I'm figuring something like 1 to 2 ohms at most. Could be considerably less or more, depends on the condition of the harness, connectors, and grounds.

Good info, Brad. I recently picked up a 75A alternator for the 'BB', and purchased one of Pete's 160A alternator kits for the '71.

I also purchased Spoke's LED tail light assemblies and have Timothy's LED gauges for the '71 too.

I've got one more idle stability thing I'm working on. All D-Jet 914's have retarded timing at idle to lower emissions. It also lowers torque, too. I don't want to get rid of the retard completely, but I think if you shim the retard stop in the vacuum cell with the right thickness of stock, you can improve idle torque significantly and still retain good emissions to pass the tailpipe test. The improved torque should cut the idle droop under load significantly. Shouldn't take much like a few thousandths. Apparently, some distributors had adjustable retard vacuum stops, but the 205 series we have doesn't seem to have them. Might be possible to retrofit an adjustable cam stop to a 205 vacuum cell and get that function.