Creating a Switch to Shut Off Cylinders on the Fly

[SVOboy]

Originally Posted by rh77

I'm a newbie here, but I'm thinking about those mercury thermostat switches on the old-school furnaces instead of a pendulum.

Also, I've been thinking of this cylinder deactivation idea on highway cruising for a long time and didn't think to cut the power to the injectors. Any concerns on OBDII? (1998 Integra LS Auto). So I can run a switch to cut 1-3 or 2-4??? I'd love to try this. It's the latest rage in Detroit for V-8s, so why not for the DIY'r?

RH77

I think the mercury switch (having played with one in physics class), has a tendency to move around too easily and it might end up switching on and off your cylinders every second on a rough road, which might not be the best thing, -_^. In any case, yes, I believe on your engine it would be 1-3 and 2-4, since it depends on firing order you don't want it to get out of balance and all. Some wires and a switch would really be a piece of cake.

[rh77]

Originally Posted by SVOboy

I think the mercury switch (having played with one in physics class), has a tendency to move around too easily and it might end up switching on and off your cylinders every second on a rough road, which might not be the best thing, -_^. In any case, yes, I believe on your engine it would be 1-3 and 2-4, since it depends on firing order you don't want it to get out of balance and all. Some wires and a switch would really be a piece of cake.

OK so I took the plunge and setup a cylinder de-activation system. I wired cylinders 1 and 3 to be deactivated by using the common ground to each injector as a cut-off and routed a switch into the cabin. It works BUT...

*Under 4000 RPM I get a moderate vibration under engine load -- it's like an unbalanced wheel feeling. Did I pick the wrong 2 cylinders? Now that I'm probably using 60 horsepower to get the 'teg going, the automatic kicks down to 3rd gear on the highway, so I run in the 4000-5000 rpm range to maintain 70 mph on flat ground -- but there's no vibration at that RPM.

*The unbalanced feeling has me concerned a bit -- would it put too much pressure on engine mounts, rods, crankshaft journals, etc???

*So, if I run at 4500 rpm at full throttle for most of a 50-mile trip, it would defeat the purpose, right?

*I know this engine is bulletproof, but, in your opinion, is it putting unecessary strain?

Thanks all-

RH77

[SVOboy]

Interesting, I was hoping you'd do it, pictures mehbe? I'll look some stuff up right now.

[rh77]

Originally Posted by SVOboy

Interesting, I was hoping you'd do it, pictures mehbe? I'll look some stuff up right now.

Hey thanks! I'll get the camera out and post some pics here shortly. Bear in mind that it won't win any "show and shine" awards, but it works. Thanks again...

RH77

[SVOboy]

"well here's the deal. you have 4 pistons (obviously) and so two of them go down, two of them come up. however it is the middle pistons that go down and the outer pistons that come up. by doing this, the motor is balanced at all times. the combustion is being put on the middle two pistons pushing them away from the head to the bottom (BDC bottom dead center) and pushing the outer two pistons to the head to the top of the combustion chamber (TDC top dead center)

if you were to disable any of the pistons, you would want to disable one of the inner two pistons and one of the outer pistons. pistons are numbered 1,2,3,4 you could either disable pistons 1 and 3 and have pistons 2 and 4 firing. Or have pistons 4 and 2 disabled and have pistons 1 and 3 firing."

And the firing order on the gsr is 1-3-2-4, so I think you're have to kill on and three or two and four, and you should be good, it might be the auto that is killing you, but you should not be using 60 hp to stay cruising, only takes about 12, mehbe less for your car being a bit aerodynamic.

[rh77]

I did some more testing on the cylinder deactivation. At idle deactivating the cylinders results in the engine rocking on its mounts forward and backward, which is probably the same shimmy under load on the highway.

I think it would be too damaging to let the car shake that much, so here's what I've done, which may have defeated the purpose. From start, I shift down to 1st and accelerate to about 4000 RPM and deactivate the cylinders. I let it get up to 5K and flip the switch. Then I use WOT to get up to speed and leave it in 3rd on the highway, so I can keep it in that sweet spot above 3500. For some reason, the increase in RPM takes away the vibration.

Question 1: Running at higher RPMs probably uses more fuel, right? But with 2 cylinders, it would use half, theoretically?

Question 2: Honda engines love the higher rev bands, but I'm pretty sure that I'm wearing out parts sooner as they are spinning faster. I would need to change the oil on a stricter schedule too.

Question 3: Is it worth it. I haven't gone through a tank yet, so the results are pending.

Any thoughts folks. Buehler, Buehler, Buehler...

RH77

[SVOboy]

I dunno about your questions, hard to think right now that I'm son cencentrated on bypassing my car's ability to tell what gear I'
m in...but I'm uploading the pcitures and data now.

[Matt Timion]

My brain has bounced back and forth on this whole cylinder deactivation thing for a while now.

It seems that it would make sense, but it also seems that it wouldn't make sense.

for each cylinder fired your o2 sensor detects it's oxygen content to determine how much fuel is needed. Two of your cylinders have essentially a zero reading after not firing them. your ECU will then attempt to enrich your fuel to compensate.

And THEN your engine will have to work twice as hard to get the same RPMs, but it will only be using half of the fuel, so you're essentially where you left off.

I have lots of reservations about doing it myself. My car currently cannot be converted to such a system as it is Dual Point Fuel Injection.

I guess the benefits of doing thsi will be determined once you fill up again.

Then again, if my engine was rocking backand forth and it became more noticable than previously, I think I would immediately stop the experiment.

But that's just me.

[SVOboy]

And I need to reupload the data, didn't seem to work the first time through.

I can't upload the closeup now, I'll try to fix whatev problem with it later though.

[rh77]

Thanks SVOboy for hosting the Pics. As for now, I'll set the DataLogger to record O2 Sensor data to see what's going on there too. I have a 55-mile highway drive to the airport to try things out today, so I'll have more data to share tonight. As for the rocking, I'm going to keep it to a minimum by keeping the revs up.

Thanks again...

RH77