![]() |
Alternator Kill and LiFEPO4 - The definite guide
Some of you may be curious what happens when a standard lead acid battery on a diesel car is replaced with a LiFePO4 Lithium Battery...
Last year when the starter battery died (again!) I decided to use a few of my other electric car batteries to see how the thing would go. The original 60Ah Acid battery was replaced with a 12.8V 20Ah LiFePO4 Battery. The small battery has no problem restarting the car several times during a short space (I kill the engine on traffic lights) and it certainly has no issues overloading the constant voltage alternator, pulling in excess of 150A after the engine restarted. If the SoC was low the belt slipped for a few minutes. Last week I decided to grab a new alternator pulley besides the new belt. 3 days later a shorted bridge rectifier on the alternator. Well... Time to get the gear running again with some mods. What I need:
Most alternator kill designs I've seen are amateur aproaches in which the alternator field wire is simply disconected shifting all the drain to the battery. Once the alternator is allowed to start, the vehicle engine must come to a halt to kill the alternator output, so not very flexible. So how to achieve this? Well... first the original regulator must be bypassed. I always found 14.4V too high of a voltage anyway. Modern lithium batteries can be charged at 1C with only 13.6V. Lead acids that are not excessivelly discharged can charge fine at 13.8V... But the idea here is to go even lower, so that the battery is receptive to charge when the engine is not being used. This can effectivelly used to take advantage of short traffic lights. Rather than stopping the engine, use the alternator at high output and cut it completelly once the vehicle is ready to go, ensuring fast acceleration and reduced fuel usage. The same is valid for braking. This is my alternator. A 90A model removed from my electric car, that will now be used on the diesel. http://i62.tinypic.com/2i264cl.jpg http://i60.tinypic.com/2hnbzax.jpg Removed regulator/rectifier cover: http://i62.tinypic.com/2yzfskm.jpg The lower slip ring is 30% Worn at 50K Miles http://i60.tinypic.com/2mpk29i.jpg View of the built in Voltage regulator. There is a monolitic IC to control the field current. The switching device is a NPN transistor. The positive brush is tied to VCC http://i58.tinypic.com/fx5xrt.jpg http://i57.tinypic.com/30sftyx.jpg http://i59.tinypic.com/2ptasud.jpg Removed the brush conector protective sealer to expose the bush connections. Be very aware that if a regulator fails with a shorted transistor things will go incredibly wrong. There are ways arround this, but for the sake of simplicity I decided to keep the stock regulator, since my own regulation device will be set to a LOWER voltage. This means, if said device were to fail closed, the built in regulator would still regulate at 14.4VDC (CV). Under 14.4V the original regulator will feed the rotor with a 100% duty or close.The mods are as follows: The built in regulator is kept powered at all times with VCC and GND. Find the output wire that will feed the brush negative terminal. Open this terminal to kill output and close to enable alternator. This can be done in real time with no nead to stop the engine. For now, the mods are pretty simple: Cut the terminal below where it was spot welded. http://i61.tinypic.com/n35bip.jpg Now separate the regulator output from the linkage to the brush http://i62.tinypic.com/117gqk9.jpg Check both points for continuity to make sure they are indeed separed http://i62.tinypic.com/6e41fm.jpg http://i57.tinypic.com/2iqfnfq.jpg Part two of this thread will discuss a PWM regulated circuit with adjustable output voltage and current. Untill I get some time to sort that out, the alternator output can be reduced by placing a 10W, 10Ohm resistor across the now separated terminals. This ensures output power is reduced at lower revs and increases as the engine RPM goes higher. The end voltage is still regulated at 14.4 but it will work as a constant current source. During braking, the two points can be shorted togueter to generate maximun power. Note: A (fast) Freewheeling diode needs to be soldered across the brush terminals to avoid high voltage inductive transisents when the rotor power is cut and the original regulator is not directly attached to it. Happy modding! |
Very cool and interesting project. Looking forward to more.
|
This worked out as expected.
The car idles at 850RPM, and the alternator pulley has a ratio of 1:3. I used two paralleled 10Ohm, 10W resistors (total 5Ohm, 20W). With the new mod, the field receives about 7V at full load, drawing 2.5A. That's 30% of the original power Resting Battery Voltage: 13.2VDC (40%SoC) Upon Starting the engine the voltage slowly ramped over a period of 2 minutes (Actually faster than with a slipping belt) until it reached 14.1V. The car electrical system alone uses 3A. Some Load tests after the voltage reached 14.1V (Battery not fully charged):
So it's charging at 35A+ at idle (It was tested a couple years ago to provide about 80A @ 750RPM @ 12.5V. Result: Cold Battery and no more belt slippage/sluggish acceleration For alternator kill mods. rather than completely turn off the alternator, reducing the output power allows an increase in charge efficiency, with transients being delivered by the battery instead. It's more efficient than waiting for the battery to be depleted and forcing the alternator to operate at full power to top up. |
On, slow and off alternator charging mode would work for me.
|
All times are GMT -4. The time now is 04:43 AM. |
Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2025, vBulletin Solutions Inc.
Content Relevant URLs by vBSEO 3.5.2
All content copyright EcoModder.com