Archives

I'm trying to figure out a way to use my SG2 to detect when my 07 Civic's engine (the 1.8) enters it's high fuel economy mode. This mode entails a switch to a cam profile that keeps the intake valve open until a piston is well into its compression stroke (thus expelling some of the intake mixture) while keeping the throttle wide open to reduce pumping losses. Sort of a semi-Atkinson cycle mode of operation.

Using the SG2 for realtime monitoring, I was expecting to encounter a condition while I'm cruising at constant speed and providing low accelerator pedal pressure, where the LOD value is low but the TPS is high. But, under cruising conditions where I'd expect the FE mode to be active I'm seeing TPS around a normal cruising value of about 20. This is roughly the same TPS value as I encounter in 4th gear at speed, when the engine speed is higher than the 3500 RPMs at which the i-vtec system supposedly will not engage.

I *am* seeing higher MAP readings when I'm cruising at speed than I am if I put the engine under a little more load, and I'm wondering if this could be the clue I'm looking for (higher MAP implies wider throttle opening?). Today, for instance, I was seeing MAP of around 10-11 while cruising compared to 14 with the engine off, and 6-7 around slight acceleration. I'd still think the TPS should be a good indicator, though that's not what I'm seeing.

Has anyone else messed around with trying to detect FE i-vtec engagement in an 8th gen Civic?

Let me see if I can help a little bit here. I think the problem is related to the drive-by-wire setup. Apparently the SG2 is reporting pedal position rather than actual throttle position.

The MAP signal remaining high with low pedal input is a dead-on indicator that the throttle is open wide. You should see similar MAP readings if you just smash the pedal for hard acceleration.

An alternative is to wire an LED or other indicator directly to the VTEC solenoid. This is present on the intake manifold side of the engine (typically, since the valves in question are intake valves). It is a pretty standard looking solenoid that has two plugs connecting to it, usually a green one with a single wire (coil) and a gray one with two wires (oil pressure sensor). Tap the green wire and watch for 12v!

You can be sure any time VTEC solenoid is engaged that you are in high FE mode.

I would assume (could be WAY off), that the ing timing would ... retard I guess, while in vtec. Worth looking into. I know on my car (very much non vtec), the ignition timing retards a great deal while in gear with tps of 0.

Thanks for the replies!

Maxx: Interesting theory. If traffic allows, on my drive home tonight I'll monitor timing and see if I can make any good observations.

shifty: Two points:

1) Just out of curiosity, since more than engine speed and oil pressure supposedly go into the vtec engagement decision on the R18, wouldn't there have to be more than a coil and oil pressure line connected to the solenoid? Maybe a line from the ECU? I'm hoping that there turns out to be an actual OBD2 PID to determine vtec engagement (which would correspond to this line) but time will tell on that.

2) Good theory on the SG reading the APP (accelerator pedal position) instead of the TPS, but I don't think that's exactly what's happening. I haven't done any tests lately, but I believe I've used the TPS gauge in the past to monitor how hard the car's working going up hills with cruise control engaged. Since the pedal doesn't move while in cruise (thanks to the DBW setup) any change in TPS readings in response to load would indicate that while it may not be reading the actual TPS, it's definitely not reading APP. I will also double-check this on my drive home tonight.


Another possibility here is that I'm actually not getting into the FE mode, although I find that a little hard to believe. Wiring an LED to the solenoid would confirm that, I suppose, but getting to the intake side of the engine in this car is a bit of a pain (look up pictures of the engine bay to see why), and I think I'm too lazy to do it.

The "coil" is the part of the solenoid that is actuates it, so the wire connected to the coil is the wire from the ECU. The oil pressure switch on the solenoid is just on or off. VTEC will not engage if this switch isn't closed.

Typically there is a function of engine speed (RPM) vs. load that is used to determine VTEC actuation. It is probably more complicated than that on the R18 due to DBW and the Atkinson mode.

I am certain Honda would provide the VTEC status through the OBD port. Even there old OBD-1 systems allowed for easy data logging which included VTEC off / on.

It would just be a single bit to look for. :)

Oh, and the intake manifold is in the front of the engine on an R18. If you lifted the hood and removed the plastic cover over the engine, you'd immediately spot it more than likely.

Ah, I understand now. I thought you meant ignition coil, to determine engine speed. Makes more sense to me now :) Good to know the VTEC mode is probably available over OBD2.

The intake manifold is on the firewall side of the engine, not the radiator side, at least in the USDM Civic application. To even see it from the top, you need to remove the cowl cover and under-cowl panel. The process is detailed in the service manual and doesn't look too bad, but I've avoided it so far out of laziness. Involves removing a dozen plastic clips or so. We're probably talking about the same thing here, just in different terms.

Depends on the Civic.

On the R18, the cat is on the front and the intake manifold is on the back.
On the K20Z3, the cat is under the floor and the intake manifold is on the front.

Outside of the vtec-YO crowd, you are looking for low TPS and high MAP. (I'm still looking for MAF as well.)

-mr. bill

On my drive home, I monitored LOD, TPS, MAP, and ignition timing. I didn't find any interesting patterns in ignition timing. TPS does vary with load when cruise control is engaged as I mentioned earlier, so I don't think TPS is actually reporting APP. Honestly I'm not even sure if what I'm observing with MAP is legitimate or not; it's hard to tell whether the higher MAP is because of the throttle opening for FE operation or if I'm just giving it a little more gas (it seems pretty sensitive and the freeways around here aren't totally flat).

mr-bill: Of course you're right, the K20's intake/exhaust manifolds are on different sides than that of the R18. Since we've been talking about the R18 only here I didn't specify earlier. Why do you say we should be looking for low TPS, though? That certainly matches what I'm seeing but if the TPS signal is actually reporting the throttle plate position then it really seems like it should be much higher when VTEC is engaged. If you've got specific observations, what are you seeing for TPS and MAP numbers? When I think I should be in FE mode, I'm seeing about 17-20 TPS and around 10 MAP (with MAP approx 14 today with the engine off). This is at 65-70 mph, engine speed 2900-3100 RPM.

Well, you could just check wikipedia... http://en.wikipedia.org/wiki/VTEC#i-VTEC

It's a function of load and rpm.

Depends on the Civic.

On the R18, the cat is on the front and the intake manifold is on the back.
On the K20Z3, the cat is under the floor and the intake manifold is on the front.

Outside of the vtec-YO crowd, you are looking for low TPS and high MAP. (I'm still looking for MAF as well.)

-mr. bill

I had forgotten the R18 still spins clockwise... I thought Honda had "fixed" that on all their newer models. :D

Honda's technical papers clearly state that the Atkinson mode works by delaying intake valve opening / closing, decreasing the LSA (lobe separation angle). It also states that is uses the DBW to open the throttle wider during this operation.

Is there an ECONO light or indicator anywhere on the dash? A lot of other Civics had one to indicate various things...

I had a chance to do some test concerning the 1000 to 3500 rpm vtec/semi atkins head function over the holiday. As we know in the rpm range under low load. The R-18 is suppose to be operating at 1.5L and not 1.8L.

I had to make a run to the store and the car was warm. My Civic is also a 5MT so it made doing this test easy. I also used my scanguage to give me the instant mpg.

One level ground, no load, no wind and temps in the mid 50's. I drove the car at 900 to 1000 rpm in 5th gear. Just off idle. The scanguage showed 36 mpg as long as I didnt move my foot on the throttle. I then sped up a tad to 1300 rpm and held it steady. MPG was then a stable 57 mpg. This was repatable going and coming. In my mind that proves the lower end of the vtec operation. Testing the top end of the rev range could be more challenging.

psy

Found this post by psy on another forum.

I was under the assumption that while the new civics (1.8L) have two lobes, it remains on the more efficient cam lobes until about ~3500rpm. Above that it switches to the performance lobe (hence the Vtec).

With low load the Atkins cycle does kick in, however, I think there is more to it than just low load. From what I can see the only real way to tell if the atkins cycle kicks in is if your mpg read is higher than what it is coasting in neutral at the same speed. Granted it would be pretty had to determine this quickly and off the top of your head, it could be done.

I generally ust try and maintain a MPG reading above 40 when I'm maintaining a constant speed. Although the hills around the area make that difficult.

shifty: no econo light on the dash in this civic.

I'm going on a road trip tomorrow. I don't have the SG2 calibrated yet but if the conditions are good (not too much headwind) I'll see what I can tell wrt MPG, LOD, and MAP. It's been windy lately though, so I'm not holding out much hope.

Technical details of the R18. (http://asia.vtec.net/Engines/RiVTEC/index.html)

mtbiker278, each VTEC or i-VTEC implementation is very different. The three engines (http://www.vtec.net/articles/view-article?article_id=396999) in the 06+ Civic all have *very* different control over the valve train.

-mr. bill

... and the i-vtec on the dohc k24 is different still - cam phasing only but it has the same atkinson-like effect as the r18 system. i-vtec has become a sort of catchall term that doesn't mean anything specific beyond variable valve timing. I'd love to see some really good reference for manufacturer specific obd2 PIDs (certain software packages seem to have fairly comprehensive collections, but they are in proprietary formats).