1987 voyager with displacement on demand...

[Ondonti]

Quote:

Originally Posted by roguetrip This is interesting, I think looking into what chrysler did to the new hemi with being able to switch off 4 cylinders would give more insight to this. With firing order 2-4-6 that would give quite a bit of rocking for each rotation and would wear the main bearings along with the rings out a bit faster by gasoline blow by from in enriched enviroment enticed with abnormal vibration. Now if rocking and wear doesn't get ya, then a way to maybe eliminate the O2 putting the system into rich when the 1-3-5 cylinders are off, it wouldn't be cheap, but make a dual exhaust with dual O2 sensors, now the bank that gets turned of, also switch off the O2 sensor for that bank so it doesn't give a bad reading. Now all that air from the cylinders won't be interferring with the working cylinders and making the O2 run rich for the other pistons which would in turn wear rings faster and have to change oil more frequent due to blowby of gas into the system. I know lots of guys with thow flame thowers out of custom exhaust manifolds let the engine run very rich and cause those issues. Also when running a dual exhaust on half the cylinders, i'd go for really small exhaust piping for better back pressure to keep RPM's up. 1 7/8's or less would probably be idea.

Or you can just trick the ECU by building a box like ED or looking talked about tht will limit the range of the o2 sensor when you have it activated.

[m90daytona]

this is just gave me a great idea for the future when gas prices soar to $8 gallon.

take almost any car you have, and remove some of the pistons and connecting rods. if the van runs okay DRAGGING 3 dead cylindrs, imagine how it would run if it only had 3 cylinders anyway.

[WVRampage]

Sounds like a good Idea,If you are geting 35 mpg it must be working.

[phantomrt]

Enjoy your 35 mpg minivan. Sell kits on ebay that promise 36% increase in fuel economy for $35 or so.

How are you calculating that gas mileage anyway? Long 70 mph road trips?

[RoadWarrior222]

Hi all,

I've been researching this since finding Adam's original postings about it on allpar. I've had a buttload of other stuff to catch up with with my minivan, so haven't got around to trying it myself yet. She's running nice now though, so might start playing soon.

A few points. Yes it's not as efficient as a designed in DOD, that's why he's claiming 35mpg not 52mpg. I think the O2 sensor position in the rear manifold has a lot to do with being able to "get away" with doing this without it running wayyyy too rich or throwing codes. It's positioned right where the flows join from each bank, thus I suspect that higher pressure exhaust gases from the active bank will "wrap" around it, rather than it "seeing" much plain air from the inactive bank. If it was further down the pipe this would probably not happen, as the flows would have more time to mix.

I think one of the big reasons that simpler DOD schemes are/were not pursued by manufacturers is that they only perform excellently under steady speed conditions. EPA tests are not steady speeds. Several manufacturers with sophisticated DOD systems have found that under EPA highway test conditions they could only get the DOD active around 10% of the time. Hence DOD systems are not "worth it" for the added complexity when they only add 1mpg to the numbers that everyone refers to. Driveability issues are also quite important in selling cars, no dealer wants a car on his lot that feels like a slug hauling a travel trailer at highway speeds when the DOD kicks in, they won't sell, customers will walk after the test drive. Ergo, smart drivers who understand the compromises involved can choose to have way better MPG than the factory can put into a car that dealers have to sell in the first place to people who probably aren't buying a car for MPG anyway.

As some of you might remember I have a spare 3.0 to rebuild and have been considering that build with this in mind. I was asking about dual throttle bodies on here, because in addition to killing the injectors, I was thinking that independant aspiration of each bank could improve efficiency further. The build is also going to address parasitic drag concerns in the motor, such that 3 cylinders will not have to work so hard to drag it along.

I do think switching banks from time to time is wise, the concerns I have about having one bank dead for too long is actually twofold, one that it will cool off too much and you get the equivalent of 10 or so cold starts worth of wear per trip as you kick the DOD in and out, and two that oil slowly builds up in the cylinder and you get overpressure when you next fire it, or maybe eventually bend or break a rod if you go hours without firing it. However, running 25% synthetic oil will help a lot with the first issue wrt wear. The second issue could be addressed by remembering to flip back to 6 cyl for a few mins in the half hour while on long trips.

Another thing that seems likely from looking into this is that I think it will only work in simple form on '87 '88 and '89 electronics, I began to suspect this sometime last year and forget what the reasons were. Anyone with '90 up can feel free to prove me wrong though.

Anyway, good luck with your further experimentation Adam and thanks for sharing you r info with us, keep us updated!

RW222

[Ondonti]

the simplest thing I would do is just have a switch that also cuts off the o2 feedback.

It will run a little bit richer then normal but once you turn off 3 cylinders it wont try to add fuel.

So DOD on a SBEC car would probably require a switch for every injector???

[RoadWarrior222]

Maybe that was it, was also getting the idea that SBECs could tell more readily when an injector wasn't firing, and they'd pull the timing way back and start throwing lots of codes.

Remembered a couple more things from going into this before...

Older O2 sensors might be more tolerant of this due to slower slew rate. A shiny new one might be fast enough to report "friggin' rich!" "friggin' lean!" alternately which could well lead to the SMEC or SBEC freaking out. Though thinking about it, that's what narrowband is supposed to do, and the ECU is meant to keep it cycling between rich and lean...

It might be wise to attempt to restrict PCV fumes to the running cylinders only, as well as try to prevent "charge robbing". If small amounts of combustible fumes or mixture get into the non-running bank, you could have some nasty superlean detonation.

I had an idea for some hairy O2 fudgination should it prove to be a problem... that of hacking up a reference voltage from a commonly available cellphone charger, with various diodes in series to drop from 3.6 or 2.4 to around the .4 to .5 voltage mark, switched in when necessary or wired to a multipole switch with the cylinder cut out.

[Ondonti]

actually super lean mixtures dont detonate. I dont think the PCV would be a real serious issue. You still have your plugs sparking. The PCV would probably just add to the potencial to foul plugs. Reason to remember to clear your plugs every once and a while when driving.

SBEC II's only run code 27 for the injector circuit being messed up when you unplug the entire harness. You can remove injector plugs and the car still runs but I obviously have not tried this while being hooked up to a scanner.

[Ondonti]

Can you buy a switch that will seperately cut 3 wires with one throw? having 6 switches (or 3 for just 1 bank) would suck.

[RoadWarrior222]

Those would be treble pole, single throw switches.
This would work though, a double throw version, just need to leave one side disconnected.
3PDT ON-OFF-ON TOGGLE SWITCH | All Electronics Corp - Parts, Supplies and Components

However, I wouldn't want my injector signals running all through my dashboard, so I'd probably put a treble pole, or a bank of single pole relays, nearer the injector harness, then wire that to a single pole single throw switch on the dash, to switch 12V to them and activate them. Then if you wanted to add stuff to switch in at the same time, you wouldn't need to tear your dash up again and fit a different switch.

[86LaserXE-XT]

hmm interesting idea, I would think a few things would be bad for the engine.
and your milage in theroy shouldn't increase since when you switch off half the engine.
you've just increased your pumping losses big time. and by having to put your foot to the floor and the computer sensing a lean mixture that you'd be pumping more fuel in than if it was just running on all 6 cylinders still with your foot barely on the gas.
only way I could think to overcome the problems is to have a system that could hold the exhaust valves open. preventing a major portion of the pumping losses from the dead cylinders. also when your running on 3 clylinders and your foot to the floor. your increasing the pressure in those clylinders dramaticly and it will wear much faster since your putting more of a load on them then if you had all 6 clylinders firing and drove WOT all the time.

as far as fooling the 02 sensor, I'd think a programable IC such as an OOOPIC(oopic.com)would be ideal. by using a virtual circuit you could send any reference voltage you wanted to the ECM and it could vary off what the 02 is reading at the time. it would simply be compensation for the overly rich readings but would know it with an input signal from your control, the possibility would be endless with the programable IC as it could read your TPS signal and cut the cylinders in and out as you wanted. it could also switch the banks that it disables so you never run one bank to long without firing.
I'd attempt to wire one up myself right now.
but I don't have any 6 banger minivans just two 2.5 powered ones. and I don't think the wife would be happy with me taking wire cutters and a welder to her 02 caravan lol.
i'm gonna have to think about this idea alot more.
thanks for the idea :)

[Ondonti]

Im quite sure he doesnt have to floor it to keep up speed.


I dont think DoD cars try to mechanically stop pumping losses either............but they still gain mileage.

[86LaserXE-XT]

ok I missread his first post, he only has to keep it floored when at low speeds :)

and I do belive this could be how they deal with pumping losses
Dodge VDS
Chevy AFM
since I can't built that setup in my garage I guess best I can do is figure out how to use my PIC to sense the TPS and control the injector shutdown and mask the 02 reading.

[comp]

interesting read ,,,any updates or wiring diagrams ???

[v6camrydriver]

I’ve got a ’94 V6 Toyota Camry (3VZ-FE) and am thinking of developing a similar system.
My current average is 10L/100Km (23.5 Mpg).
Firing order is 1-2-3-4-5-6.
I will be disabling cylinders 2, 4 and 6.
It uses sequential injector firing but all the injectors on each bank share the same supply wire (each has a separate wire to the ECU which then grounds each one in time) so wiring in a switch will be easy.
It also has 2 oxygen sensors, one for each bank. OX1 senses cylinders 2, 4 and 6 while OX2 senses cylinders 1, 3 and 5. My solution for fooling the ECU is to permanently wire the OX2 sensor into the ECU inputs for both OX1 and OX2, hopefully tricking the ECU into thinking everything’s running as normal and not dumping in extra fuel.
A few things I’m worried about are extra oil consumption, plug fouling, damaging detonation of PCV gases and EVAP gases, and bearing damage.
I’ve been reading up on Honda’s VCM system (honda.co.nz/hondaV6/about-vcm.html), which tells me these problems must be solvable.
Any thoughts/ideas/suggestions.