Digital EFIE operation
Digital EFIE operation – Previous EFIE Designs First, lets have a look at how oxygen sensors work. Have a look at Figure A below. Here we have a graph that is a representation of the voltage output of a typical oxygen sensor while the engine is running. Note, that this is only an approximation of a real voltage graph. A real graph would be much more jagged and would not be so regular as this one. But I’m using this graph to make it easier to visualize the concept of what the sensor is doing.
Narrow band oxygen sensors don’t tell the ECU what the air/fuel ratio is. They only tell if the mixture is rich or lean. The line that is marked “.45” volts denotes the make/break point for the sensor’s voltage output. Any voltages that are higher than .45 volts is considered to be rich, and any voltages that are less than .45 volts is considered to be lean. When the sensor produces .45 volts, that is considered to be the correct air/fuel mixture which happens to be 14.7 to 1, air to fuel (by weight). The trouble with narrow band sensors is that they can’t tell the ECU how rich or how lean the mix is. They only tell the ECU “rich” or “lean”. Therefore, in normal operation, they are constantly changing voltages similarly to the graph in Figure A.
Now look at Figure B. The blue line in this graph represents how an EFIE changes the voltage graph of the sensor. As the sensor produces its voltages (as represented by the red graph), the EFIE adds additional voltage. We are showing an EFIE set to 350 millivolts (.35 volts). Therefore the output of the EFIE that goes to the computer will be the voltages in the blue line on the graph. Because higher voltages mean a richer mix to the ECU, the ECU will then lean the mix when it “sees” these “richer” mixture signals coming from the oxygen sensor.
Almost all EFIE designs that are in use today work like the above graph, by adding a voltage to the output of the oxygen sensor. While this approach does work, and has been the only solution available for many years, it has 2 problems that make it not the ideal design.
1. There is a definite limit to the amount of voltage you can add. Notice that if we added .5 volts in the above graph, that the blue line would never dip below the .45 volt line. This is an illegal condition and the ECU will quickly stop using the oxygen sensor if it never sees the voltage transitioning from rich to lean. In actual fact many ECUs need to see voltages lower than .45 volts before it will consider that the mix is lean, and so often you can’t set an EFIE higher than 250 millivolts or so without throwing engine error codes.
2. It takes a relatively large change in the voltage to make a small change in the air/fuel ratio. This wouldn’t be a problem in itself, but coupled with the fact that we can only add a limited amount of voltage, this causes an end result of a small change in air/fuel ratio.
There is one other approach in EFIE design in use today, and that is to use an amplifier. Instead of adding voltage to the sensor’s output, EFIEs of this type will amplify the signal. This, in effect, multiplies the signal. This is a better approach in that the lower voltages are not increased as much as the higher voltages, and you should be able to shift the air/fuel ratio further than with a voltage “adder”. However, it is still limited to the amount it can shift the voltage before all voltages are higher than .45 volts. Also, the amplified voltages at the top of the graph can get quite high, possibly high enough that it will set off alarms in the ECU.
Enter the Digital Narrow Band EFIE
There are other EFIE designs being marketed as “digital”. In each case, as of this writing, the only thing digital about them is the pot used to control the EFIE. It’s a digital pot and will have one of 64 or 128 resistance values, or possibly more depending on the resistor chip design. While this is cool, it makes no difference in the operation of the EFIE. It will still be operating like one of those described in the section above.
Our new Digital Narrow Band EFIE operates completely differently from any other EFIE made. Our new EFIE is called digital, because it’s output is either on or off. Or in other words is either high or low. Or to put in terms the ECU will understand, the output will be either rich or lean. Or to put it in terms of voltage, the output is either going to be .100 volts or .900 volts. This is perfectly acceptable to the ECU and tells it exactly what we want it to see. But because it’s output is only one of 2 states, we rightfully call this device a “digital” device.
So how do we know when to switch from the high state to the low state? We have a comparator in the EFIE that “decides” when to switch states. If the EFIE were to be set so that there was no change in air/fuel ratio, the comparator would be set to .45 volts. This would mean that if the voltage coming in from the sensor were below .45 volts, the output would be low, and likewise if the voltage coming in from the sensor were above .45 volts, the output would be set to high. This would cause a flat response in the ECU where it would provide the same air/fuel ratio as if the EFIE were not involved.
To lower the air/fuel ratio we need to make the mix appear richer. In order to do this, we make the EFIE transition to a high output even though the input is below .45 volts. In other words, instead of using .45 volts as the switching threshold, we use .20 volts (see Figure C).
By adjusting the pot
By adjusting the pot on our new EFIE, we are adjusting at which voltage the comparator will use to determine if the output should be set to high or low. In the graph below, we show 2 comparator voltages for comparison. At .45 volts, we can see that the output will be high about 1/2 of the time. This is the same as it would be without the EFIE. Now notice the line at .2 volts. By setting the EFIE’s comparator at .2 volts, the EFIE output will be low for about 30% of the time and high about 70% of the time. This will make the air/fuel mix look richer than it is, and the ECU will respond by leaning out the mix.
Note that .2 volts is probably too low for your vehicle. You will probably not need to set it this low. We only set it here to make it easy to see the principal involved with our new Digital EFIE. An actual setting would probably be closer to .300 – .325 volts.
Note: When downstream sensors need to be treated, do not use this device. Use an older style, voltage adding type of EFIE. The reason for this is that we’re not certain how the downstream sensor information is used by the ECU. In some cases, we have read the voltages from downstream sensors and they don’t jump up and down as shown in the graphs above. We’ve seen them just float around in the .2 to .3 volt range, not changing much. This is not the behavior that the Digital EFIE was designed for. It may work fine. But we prefer that the ECU just see the same behavior, but shifted up a bit, the way a voltage adding type of EFIE will do. Any of our Narrow Band EFIEs that aren’t labeled “Digital” will work for this application.
Using this device, some people have been able to lean the mix to the point that the engine will die. However, in some cases, it is still necessary to do other treatments to get the leaning results needed. For instance many ECUs use the downstream sensors as part of the air/fuel calcs, and many more will use the downstream sensors to verify the upstream sensors and throw odd engine errors. In these cases, downstream EFIEs are needed to get the needed results. That’s why we created the Digital EFIE & MAP/MAF Combo It has 2 digital EFIEs for the upstream sensors and 2 analog EFIEs for the downstream sensors. This will give you the optimum treatment for each sensor, and is the most powerful solution we’ve seen yet for optimizing your engine for use with HHO or other fuel combustion enhancement technologies.Tags: electronic fuel enhancer, Figure A, figure B, fuel cell, hho, hydrogen, hydrogen generator, MAF sensor, water for hydrogen