$10 DIY cylinder management for honda
 

This will only work if your o2 sensor only reads some of the cylinders

Awesome news ! My DIY cylinder management system actually works!

This is my project thread http://ecomodder.com/forum/showthrea...-si-24585.html


Variable displacement experiment. 91 Civic Si B18 - YouTube

The first thing that I did was checked the resistance of my injectors. They were all between 2.4 and 2.6 ohms. (92 - 95 should be around 8 ohms FYI) Being that I could not find any 2.5 Ohm resistors that could handle more than 1/4 watt of power I wired four 1/2 watt 10 Ohm resistors in parallel, which yields a final resistance of 2.5 ohms.

http://i1339.photobucket.com/albums/...psb0f0ae12.jpg

http://i1339.photobucket.com/albums/...ps5fc499eb.jpg

The next thing that I did was remove the plug from injector one and unplugged the two plugs in the wiring harness that go to the injectors. (see next picture)
The green one goes to the resistor box (nonexistent in 92-95 Hondas) the large white plug is where you test what wire goes to where. They are also color coded, I always check with a volt meter to be certain before I go cutting the wiring harness.

http://i1339.photobucket.com/albums/...ps0b5ad4e2.jpg

To make sure that this would fool the computer, and to prevent myself from wasting a few hours of wire in something that would not work; I simply put the resistor directly on the cut wires and completely bypass the injector.

http://i1339.photobucket.com/albums/...ps45054d59.jpg

I went in the car turned the key and to my amazement it started with no check engine light!!!

Granted, it did run a little rough or in just three cylinders Lol..
Now, it was just time to wire it all in!
To make the alternate circuit I used a DPDT switch from radio shack (On-Off-On) I was hoping to find a DPST On-On switch, but it is just as easy to skip the center off position.

http://i1339.photobucket.com/albums/...ps5659178a.jpg

On this style switch there are six terminals. The three on the left and the three on the right that are completely independent of one another(one circuit for each injector)

On the upper two terminals I connected the resistors and spliced into the constant power to injectors 1 & 4
The middle terminal I connected to the ecu side of the injector ground wire (brown in the first picture)
And on the lower terminals connected to the injector side of the ground wire (brown in the first picture)
As a side note the wire color for cylinder number four is yellow.

http://i1339.photobucket.com/albums/...ps229c2251.jpg

Listed tools and items needed
• Resistors equal to resistance of injectors (2.5 ohm in 88-91 civic 8 ohm in 92-95) ~$4*
links
NTE HW010-10 - 1/2W 10 Ohm 2% Resistor : Resistors | RadioShack.com
8-Ohm Non-Inductive Resistor : Resistors | RadioShack.com
• Soldering iron with Solder
• Good wire (12-18 gauge, I used about 30 feet....) ~$6*
• Heat shrink ~$4*
• Wire cutters/ strippers
• 1 DPDT On-Off-On switch (Or a DPST On-On switch) ~$4*
* cost is assumed if buying new.

Most people who do wiring should have all of these with the exception to the switch and possibly resistors, all I purchase were the resistors and the switch.

This will only work if your o2 sensor only reads some of the cylinders

Let's see some fuel economy improvement figures from your cylinder management circuit.

WTH is this?

It's also called "shutting off said fuel injectors" without also "deactivating the related intake and exhaust valves."

It's called not looking at existing data. :/

As if no CEL means it works.

There will be pumping losses, but I will be posting mileage numbers as I get them.

Should probably have done that first.

Hmm... I'll play devils advocate here, probably because I haven't seen the existing data yet, but this makes sense to me. If we're only running on two cylinders, we only make half the power, and can run the motor much closer to WOT for the same fuel consumed and power outputted. This means reduced pumping losses.

Why won't this work? ;)

If this setup alternated between cylinders 1,3 and 2,4 ever few seconds, it could probably lower overall engine temps and reduce cooling needs. I remember reading about how the Cadillac Northstar cars could run w/o any coolant by cycling off 1/2 of the cylinders at a time and using a bit of fuel to cool the ones that weren't firing; sort of a keep-alive mode until the car made it to the next shop.

Just look up the cylinder deactivation threads- they're long and exceedingly involved, why do it all again?

Sorry, installed everything, drove it, and posted it on here.

a8ksh4, That would be much more involved.. tricking the ecu to say injectors are firing is easy, that would involve fooling the o2 sensor.

I do know that this has been brought up many times before, but I have addressed every problem brought up in other threads. (with the exception of pumping losses) are there any other problem you can think of?

You haven't addressed anything except the CEL, and a piece of tape over it would take care of that.

Loock at this page :
Traction Control - How does TC work?

Racelogic is making a traction controle system that cut a part of the power of the engine by disabling a cylinder, and alterning that cylinder so the engine stay in a steady thermal state.

Here is a part of the text:
"Racelogic Traction Control prevents lean burn by removing 100% of the pulsed fuel delivery – essentially the affected cylinder takes a gulp of fresh air; the in-cylinder temperature remains virtually unaffected.

Prolonged fuel cut on one particular cylinder would cause scavenging of the petrol lining the inlet tracts, and when the next full fuel pulse arrived, it would be partially reduced in quantity by the re-wetting of these tracts. RL Traction Control rotates cylinder cutting to prevent this situation from occurring."

It would be perfect to be able to set a dial to cut a % of the fuel. And since it's by pulse, it would be possible to cut different % that 25%(1 cylinder) or 50%(2 cylinder)

ive often thought about deactivating a cylinder when at idle at a stop light for more than 3 seconds(and only then, NOT driving). i havent used EOC in a while and i dont shut my engine off at lights. i wonder if this way would be usefull. just flip a switch and it keeps running on 3 cylinders, then when you are ready to pull out flip the switch back on and off you go. i dont like the idea of cylinder deactivation while driving.

Pumping losses are the major problem that has to be addressed. Every other problem is just icing on the cake.

Data please.

Its only a 10 min job to pull the valve cover and pull the rocker arms... lol
( I don't plan on doing that)

sounds about what my sidekick did with a near dead hole. 'had 105~ compression'
best i could muster was 22.5 running on 3 of 4.

now that i think about it i should have pulled the power to that injector and would have still gotten the EPA rated mpg number 26. with a near 50% loss in power.:snail:

Frank's pooh-poohing aside, I still want to see how this works out. I haven't seen all the data, read all the threads, etc. There aren't enough hours in the day, and this is the thread I saw. So keep at it and tell us what you discover. If it simply proves the data as established is right, then no worries. If it comes out you experience a significant improvement, that's interesting and I want to know more.

I have always planned on A-B-A testing as soon as I can for data.

For all of those people here who are considering/performed the disabling of one or more fuel injectors to kill pistons to achieve cylinder deactivation:

The dead pistons are still passing air, which still requires work to be consumed. This is work that would otherwise have been used to push the vehicle forward.

Basically, you're burning gasoline to pass air through an engine, and you'd be very lucky to get the same EPA figures for running an engine with dead cylinders, as you would with a properly running engine.

Yes that's understood from multiple previous threads but what about only doing it at idle once to operating temp? I'll answer my own question by trying it tomorrow when I get to work. I have an afr gauge in my car so it should be easy to tell. Ill take a pic of the afr at idle at operating temp and then unplug an injector and take a pic and see what happens

You mean, instead of having the engine just power the accessories (like power steering, alternator, water pump, and A/C), have the engine power these accessories AND also pump air through itself?

Just an observation - the O2 sensor is monitoring the exhaust in the manifold, and without the injector firing it will read an excess of oxygen and try to richen the mixture. I cannot imagine how this will save fuel or improve mileage especially with only one dead cylinder unbalancing the firing pattern.

I try to save fuel by deactivating all of my cylinders.

UFO is correct, the only important thing in this topic is what the computer ACTUALLY does when it sees a dead hole. not every car is the same, not every ecu prom is written the same, not every ecu has the same fail safe modes, etc. it is important that we not totally bash an idea because of a hunch we have that it will work or not. especially when no tests have been done to prove the otherwise.

that being said i have no experience in my personal vehicles type of tuning software, im sure others could explain it better. all i can do is post my findings in the mosts scientific way i can, using an afr gauge and monitoring the ACTUAL mixture coming out of my exhaust.

this morning when i got to work i wanted to pull into the lot and recreate a stop light. i pulled in and took a picture of my afr after 5,10, and 20 seconds of running. the afr bounced between 14.5-14.8 and i will include 3 pictures of that. then i walked out and pulled the number 1 injector off. with the notable misfire, the afr proceded to bounce between 17.7-17.9 at the same 5,10, and 20 second intervals. i will include 3 pictures of that. the ecu did not compensate the mixture it simply ran leaner. the only downside is the engine light came on, which i knew would happen. stacy can test this stuff easier because he already has the resistor wired in to prevent the light from coming on.

if the computer doesnt set the light on during the misfire and the same 17:1 afr is coming out of the tailpipe then it might be worth looking into more. this morning was a crap shoot, it couldve went either way. just so happens my ecu reacted in this manner. its not safe to say every other one will do the same

http://i1050.photobucket.com/albums/...psabcbc0ab.jpg
http://i1050.photobucket.com/albums/...pse6158807.jpg
http://i1050.photobucket.com/albums/...ps97e99480.jpg
http://i1050.photobucket.com/albums/...ps5d190a80.jpg
http://i1050.photobucket.com/albums/...ps3a08d09b.jpg
http://i1050.photobucket.com/albums/...ps5a9363d7.jpg

^^ You're killing me.

The only way I could do that would be to 1) drill out the little anti-tamper plug that blocks off the adjusting head and 2) attach a long cable by which I could adjust my mix on the fly at the carb.

For some reason I don't think Sweetie would smile on that. She's pretty accommodating about some things, but there's a limit.

I want to save a lot of money on fuel, but I understand divorce is expensive too.

Unless you have two O2 sensors one for each bank more common on v8s and some v6s, the ECU will see lean and dump more fuel to get it to 14.7. If you have 2 O2s then you just need to send the signal from the one running all cylinders back as both sensors that will be a bit of work or you could create a fake O2 like is used for off road cars with the cat removed to eliminate the CEL.

What would be easier/better for the engine is to custom tune the engine to run leaner when the reading the idle section of the map, if the ECM allows for it.

...a case of "unintended consequences" arising from "uneducated tinkering."

There's no way to deactivate just one cylinder on a carbed engine via the fuel system. By playing with fuel mixture on the fly you would be making the entire engine run leaner

Unless I'm remembering some other thread, the OP has a header and the O2 sensor is only installed on one of the exhaust pipes, so it is only fed by a single cylinder.

Measuring AFR won't tell you anything about how much fuel is being burned, unless you also know how much air is coming in. And measuring it on a single cylinder's exhaust when you are disabling a different cylinder tells you even less about overall fuel consumption.

If you have an MPGuino, you can get a reasonable idea of how much fuel is being burned. You need to tap into the signal for a non-kill injector, then multiply the fuel usage by 0.75 because only three injectors are working when the one cylinder is killed. Chances are that the MPGuino raw readout will go up, but if it does so by less than 1/3, then you have a fuel savings.

The SG and UG options might be quite inaccurate for this usage, because they estimate fuel used from a bunch of other parameters. And shutting down one cylinder violates at least some of the assumptions made for this calculation, rendering is much less reliable as an estimate of fuel usage.

-soD

I'm running a vx swap and my afr gauge is reading all 4 cylinders. Its tapped into the lower side of the converter. So the 17:1 ratio is an average of those 4. The 3 that are running are still probably around 14.7 but the one that's pumping oxygen in with 0 hydrocarbon is changing the average to lean.

Remember I only tried this at idle because there are less variables, there's no change in tps, map, iat, egr. So you can expect the same result every time you do it. When ur driving everything changes and has a different fuel trim relating to what throttle position, load, coolant temp. Etc so its unpredictable.

I'm curious to see what happens when Stacy gets it going, who knows maybe itll work.

If you wanted to get involved in this, to take care of the pumping losses, you could have someone machine a spacer with another throttle plate in it on teh two cylinders you deactivate typically. Wire a solenoid into the same switch so that when the cylinder injectors are inactive, it also kicks the throttle plate closed for those two cylinders.

You'll still lose some from the cylinder pulling vacuum against those plates, as well as friction losses from driving the valves still, but you won't be sucking/squeezing/[not]banging/blowing, anyway.

Might wanna do the exhausts too. Aw heck, just do the regular valves.

I was thinking that plugging the intakes off would basically negate any need to do the exhausts... since nothing would be flowing through anyway. The only time I've ever actually tried it was on a bench 2cyl 2 stroke though. If I capped one carb and removed the fuel line, nothing would come out of the exhaust on that side and it was obviously running crappy from only having one cylinder while still spinning two.

I'm trying to imagine the flows w/an intake butterfly; seems to me if a valve is open and the piston is moving, flow is gonna happen even if it is mainly just in-n-out through the exhaust valve.

Exhaust valve is only open on the up stroke... if there's nothing to push, nothing is gonna come out.

There'll be like an atmosphere of vacuum against the new throttle plate in the spacer, so /some/ flow is going to still get by, but it's gona be far reduced from original flow.

The intake, compression and power strokes will happen with a vacuum, but towards the end of the power stroke, when the exhaust valve opens, gas would rush into the empty cylinder. The last upstroke, would then be done with pressure acting on the piston.

Cancelling one up and one down stroke with vacuum, leaves us with one down stroke with vacuum and one up stroke with pressure, netting a significant loss.

I was going to use butterfly valves on the intake side of a V8 engine, in my own cylinder deactivation idea, which is documented here. This was a project supported more by wishful thinking than by thermodynamics analysis, which I could have done even then, but didn't.

However, due to the reason that jakobnev just mentioned, I abandoned this idea. Have you ever tried to pull on a piston to create vacuum? It's pretty frickin' hard to do.