Honda Electrical Load Detection (ELD) Bypass
#201
When I built my F20B/T2T4 powered AEM accord..
My Accord build carputer communicating with ecu REAL TIME TUNING - MP3Car.com
..I spent just under $500 building an alternator for it. And I am as deep in the business as anybody could ever get. My build needed to produce 100amps at idle. This was difficult but attainable. It involved all Toyota electronics in my Honda alternator and a custom made smaller diameter pulley that made the alternator spin faster. Aside from the much greater output at any rpm this setup had no ecu control. For me this was no big deal because I was wiring in a custom and completely universal engine controller. The rest of you might have to deal with the dreaded check engine light
My Accord build carputer communicating with ecu REAL TIME TUNING - MP3Car.com
..I spent just under $500 building an alternator for it. And I am as deep in the business as anybody could ever get. My build needed to produce 100amps at idle. This was difficult but attainable. It involved all Toyota electronics in my Honda alternator and a custom made smaller diameter pulley that made the alternator spin faster. Aside from the much greater output at any rpm this setup had no ecu control. For me this was no big deal because I was wiring in a custom and completely universal engine controller. The rest of you might have to deal with the dreaded check engine light
#202
It's been a while since the last posting to this thread, and I would like to revisit it. I actually have a 2012 Civic LX sedan, not an Accord, but it appears the ELD situation is the same for most US Hondas. I make a lot of short hop trips, and my battery just isn't kept charged by the Honda charging system. The only way I can keep the alternator voltage at 14.4V so that the battery will charge is to drive with the headlights on. But that wastes power, so the original posts by keep_hope_alive are very interesting to me. But I want to describe doing the bypass a different way, and see what everybody thinks.
The KHA method cuts the line going from the computer to the ELD, and runs it through a switch that either makes a normal connection to the ELD or connects it to ground through an 820-ohm resistor. But I wonder if it would also work to leave all the lines as they are, but have the switch connect the ELD line in parallel to ground through that resistor. Since the ELD is still connected, and pulling down the line voltage to some extent on its own, the added resistor to ground could probably be a higher value than 820 ohms and still keep the voltage below 2.1V.
Alternatively, instead of a resistor, a 2V zener diode could be used which would conduct at any voltage above 2V, and keep the voltage at that level, but have no effect if the ELD takes it below that voltage on its own. A 2V zener isn't a common item, so maybe three forward biased regular diodes in series would produce close enough to a 2V drop to work. They have a forward drop of about .6V each.
The other thing I'm wondering about is whether all of this could be done inside the passenger compartment by locating the ELD wire and a ground wire somewhere near the computer, attach clip-on connectors to them, and bring those new lines up to a simple SPST switch under the dash, with the resistor or diode(s) soldered to one switch terminal. If the switch is open, then the original circuit would operate normally. When it's closed, the additional path to ground would be connected in parallel with the original circuit.
I should add that my 2012 Civic does indeed have an ELD. I don't know about later models. But what I'm missing is information on the ELD line. I saw somethere that it's pin 15 on the E connector, and red/green. But I don't know if that's correctfor my model. Where does someone go to get that information for his model car? Also, is it possible to get to the connector, or to the wires involved, under the dash, or is this idea just impractical?
Any comments or suggestions would be appreciated.
Edit: It appears the ECM of my car is not inside the passenger compartment as I thought, but in the engine compartment next to the fuse box. And it's not clear how I would do any version of an ELD bypass. It looks like the fuse box is connected to a large plastic pipe that goes through the firewall, and I don't see any slack that would allow removing the box. And that would have to come out to get access to the connectors or wires at the ECM. So I may be dead in the water on this.
The KHA method cuts the line going from the computer to the ELD, and runs it through a switch that either makes a normal connection to the ELD or connects it to ground through an 820-ohm resistor. But I wonder if it would also work to leave all the lines as they are, but have the switch connect the ELD line in parallel to ground through that resistor. Since the ELD is still connected, and pulling down the line voltage to some extent on its own, the added resistor to ground could probably be a higher value than 820 ohms and still keep the voltage below 2.1V.
Alternatively, instead of a resistor, a 2V zener diode could be used which would conduct at any voltage above 2V, and keep the voltage at that level, but have no effect if the ELD takes it below that voltage on its own. A 2V zener isn't a common item, so maybe three forward biased regular diodes in series would produce close enough to a 2V drop to work. They have a forward drop of about .6V each.
The other thing I'm wondering about is whether all of this could be done inside the passenger compartment by locating the ELD wire and a ground wire somewhere near the computer, attach clip-on connectors to them, and bring those new lines up to a simple SPST switch under the dash, with the resistor or diode(s) soldered to one switch terminal. If the switch is open, then the original circuit would operate normally. When it's closed, the additional path to ground would be connected in parallel with the original circuit.
I should add that my 2012 Civic does indeed have an ELD. I don't know about later models. But what I'm missing is information on the ELD line. I saw somethere that it's pin 15 on the E connector, and red/green. But I don't know if that's correctfor my model. Where does someone go to get that information for his model car? Also, is it possible to get to the connector, or to the wires involved, under the dash, or is this idea just impractical?
Any comments or suggestions would be appreciated.
Edit: It appears the ECM of my car is not inside the passenger compartment as I thought, but in the engine compartment next to the fuse box. And it's not clear how I would do any version of an ELD bypass. It looks like the fuse box is connected to a large plastic pipe that goes through the firewall, and I don't see any slack that would allow removing the box. And that would have to come out to get access to the connectors or wires at the ECM. So I may be dead in the water on this.
I've used zener diodes in simple regulated power supplies before. We get back to current draw and we'd still need a resistor to not provide a short circuit but we'd also need to be concerned about current draw.
The idea of presenting a fixed voltage drop to the ECU is of interest to me for the newer cars - which include a circuit board with a shunt resistor located on the battery negative terminal. that circuit communicates with the ECU via LINBUS so my old process isn't compatible with new models (2013 and newer). I now need to figure out how to trick the circuit board into seeing a fixed voltage drop in place of a shunt resistor (which i think the ELD actually is). I've taken one of these battery monitors apart and it's a pain and very destructive.
#203
thanks for the interest and ideas. the primary reason for my original recommendation was to develop a plan that would be safe for the ECU and something most people could attempt. My concern with running parallel loads is the ability for the ECU to source enough current. While we are indeed interested in voltage drop, we must also be concerned with current draw of the load (or combined loads). I didn't feel like chancing a damaged ECU with a parallel load.
I've used zener diodes in simple regulated power supplies before. We get back to current draw and we'd still need a resistor to not provide a short circuit but we'd also need to be concerned about current draw.
The idea of presenting a fixed voltage drop to the ECU is of interest to me for the newer cars - which include a circuit board with a shunt resistor located on the battery negative terminal. that circuit communicates with the ECU via LINBUS so my old process isn't compatible with new models (2013 and newer). I now need to figure out how to trick the circuit board into seeing a fixed voltage drop in place of a shunt resistor (which i think the ELD actually is). I've taken one of these battery monitors apart and it's a pain and very destructive.
I've used zener diodes in simple regulated power supplies before. We get back to current draw and we'd still need a resistor to not provide a short circuit but we'd also need to be concerned about current draw.
The idea of presenting a fixed voltage drop to the ECU is of interest to me for the newer cars - which include a circuit board with a shunt resistor located on the battery negative terminal. that circuit communicates with the ECU via LINBUS so my old process isn't compatible with new models (2013 and newer). I now need to figure out how to trick the circuit board into seeing a fixed voltage drop in place of a shunt resistor (which i think the ELD actually is). I've taken one of these battery monitors apart and it's a pain and very destructive.
In any case, yesterday I found that the small two-wire connector that connects to a strange device on the negative battery terminal was apparently not making good contact, and hasn't been for five years. Disconnecting that and reconnecting it has changed things for the better. The low voltage alternator output now stays in the 12.8 - 13.2V range, rather than 12.3V. So that may be good enough that I won't have to do the ELD bypass. But it will take a few days to determine that.
Anyway, thanks for your response.
http://opensiuc.lib.siu.edu/cgi/view...text=auto_pres
#204
I guess I see the current situation a bit differently. I found a pdf that describes the Honda dual-mode system, and it shows the computer's ELD line as being an input with a pull-up resistor to 5V. If that's correct, then all the ELD itself does is sink current in proportion to the current drawn by the system, so adding a parallel path to ground would only reduce the sinking that the ELD has to do - and lighten its load. And the ECU isn't really sourcing current at all, at least not in the sense of a low-impedance amplified source.
Last edited by keep_hope_alive; 12-19-2016 at 08:57 PM.
#206
i suspect the solution is in the battery monitor, but i need to study some more schematics to figure out why the headlights trigger it. Maybe we can just "trick" the circuit into thinking the head lights are on?
#209
KHA, I have read this thread for the first time yesterday. I have read many other forums as well. I hope you do continue to post your results.
My interest is a bit different than some on here as I am using an R18 engine on a test stand with an aftermarket ECU (SDSEFI). So my interest is what happens when many of the ECU (systems) wires are not connected, like the ELD and LIN terminal on the alternator....
For the sake of sharing, and I apologize if it is a bit O.T. I just found that the 9th gen alt//reg have a default setting that allows the alternator to work like and "old" fashioned one wire system when the LIN terminal is disconnected. Cool! I can post a screen shot more on this if others want, but it may be too O.T. and I do not want to be viewed as hijacking, so I will hold off on that for now..
Thanks to all who posted on this thread - great read.
Charlie
My interest is a bit different than some on here as I am using an R18 engine on a test stand with an aftermarket ECU (SDSEFI). So my interest is what happens when many of the ECU (systems) wires are not connected, like the ELD and LIN terminal on the alternator....
For the sake of sharing, and I apologize if it is a bit O.T. I just found that the 9th gen alt//reg have a default setting that allows the alternator to work like and "old" fashioned one wire system when the LIN terminal is disconnected. Cool! I can post a screen shot more on this if others want, but it may be too O.T. and I do not want to be viewed as hijacking, so I will hold off on that for now..
Thanks to all who posted on this thread - great read.
Charlie
#210
10th Gen Civic ELD help
Has there been any updates on the newer model ELD's? I will be installing an audio system in a 10th Gen Civic next month and I am concerned with the lack of voltage, as many of you are.
There is so much good information about ELD bypass, and other things, in this thread. Thanks for all the tips, and thank you KHA for the original post and all of your hard work researching, testing, and posting here.
There is so much good information about ELD bypass, and other things, in this thread. Thanks for all the tips, and thank you KHA for the original post and all of your hard work researching, testing, and posting here.