Lowering idle RPM?
 

Hello all, I finally got Kpro for my K24A2 swapped 04 Civic, so I can now configure how the engine operates to my liking.:) Right now, I'm messing with the idle RPM. For now I set the idle to 700 RPM, which seems to actually be more like 650. Anyways, it seems quite happy at that RPM. I'm not sure how much fuel dropping the idle by ~100 RPM actually saves (if any), but I could easily test that.

My main concern is whether idling too low could in any way damage or reduce the longevity of the engine. The main ways I could see this being a problem is if the oil pressure was inadequate at a lower RPM or if the oil splash to non pressure lubricated parts like the valvetrain was significantly reduced, but then again at a lower RPM I would expect that not as much lubrication would be needed in the first place, so how much of a problem is this realistically?

Looking forward to hearing opinions and experiences on this one!

You're welcome to my K24A2 tune, if you'd like to use it as a starting point.

I found my engine got rough below ~745rpm. Granted, I had removed the balance shafts for both fuel economy and power reasons, but the Insight has no subframe, the engine is bolted directly to the chassis (with poly mounts in this case), and the stiff and lightweight aluminum chassis transmits those vibrations well, so I didn't go any lower to avoid harshness.

I don't see a low idle as hurting it. There's not much load on the engine. K's have great oil pressure, especially off VTEC, because they're designed to have extra pressure due to so much of it being bled off to feed the VVT and VTEC system, which are oil-pressure powered.

Low idle RPM shouldn't hurt it. What hurts an engine at low RPM is when you put load on it (i.e. lugging the engine). Long as it has oil pressure, you should be fine.

Cool, thanks. What mods is that tune set up for? For now I'm using the stock TSX basemap in Kmanager and as far as I can tell not being able to monitor fuel trims and air/fuel ratios since I don't have an exhaust and oxygen sensor yet it seems to be working reasonably well. I got 42.2 MPG last tank despite everything being wrong, so that's not too bad for something so quick. That should improve a bit when the brakes aren't dragging, I get a hood, exhaust, oxygen sensor and taller geared 6 speed trans, suspension is fixed and alignment is correct, etc.

Mine doesn't have balance shafts either and it still idles smooth enough at 600-700 RPM that I usually can barely feel it running over the vibrations from the road, tires, etc while I'm coasting and it never vibrates so much that it annoys me. Since I don't have an exhaust system yet, I actually prefer the lower idle because it's quieter.

As for oil pressure, I'm not sure what it is since I don't have a gauge yet. I imagine it is probably pretty good since the chain driven oil pump spins 1.6 times faster than the crankshaft. I was more concerned about inadequate oil splash at a lower RPM, but I wouldn't think 600-700 RPM would be so low that I would have to worry much. It's not like I'm trying to get it to idle ridiculously low, plenty of cars idle at that RPM with no problems.

Thanks, that's pretty much what I'm thinking. Now that I lowered the idle I don't take off or creep forwards/backwards without using the gas anymore, I always bring the RPM up to 800+ before letting the clutch out. I'm not sure taking off from low idle without bringing the RPM up is really "lugging", but it doesn't seem to like it so I don't do it.

Here is a link to my repository of tunes:

https://drive.google.com/file/d/13Dz...ew?usp=sharing

The motor was a K24A with the 04-05 intake, Accord cable throttle body, stock header, 2.5" exhaust, RSX oil pump, and 06-08 TSX cams. I had a 50° VTC with a 40° limiting pin, but it should work fine even if you have only a 25°.

The only thing I'd caution is to listen for knock, as the tunes were very much maximizing ignition advance for each octane level - the 93 tune is essentially MBT everywhere but at WOT below ~2250rpm, where I still had to pull some advance. You can, of course, simply flash a 91 tune if you're running 93 and still getting knock on the 93 tune, but I'd be happy to help adjust the ignition tables to fix that.

There are lean burn and non-lean burn variants for each fuel grade. You'll want to tick or untick the boxes based on what sensors you have - mine was complete, with upstream and downstream O2 sensors and full emissions set up. You'll also want to adjust the transmission ratio table.

42mpg is great. I found gearing was absolutely key, and gained another 5+ mpg when I got my 3.4 final drive and 9th gen 6th installed.

I don't know anything, but I'd think main concerns with low idle have to do with alternator power output, AC cooling ability, and NVH concerns.

On my motorcycle, I lower the RPM to the point that I end up stalling it too frequently, then bump it up slightly.

Awesome, thank you so much! I'll try those and see how they work for me. I have a very similar setup- JDM K24A, USDM 06 TSX cams, RSX oil pump, OEM oil cooler, Accord cable throttle body, custom cold air intake, and stock header for now although I will eventually need to change that because the collector is directly above the back of the subframe and it would be impossible to get an exhaust on it. I'll probably get a long tube 4-2-1 header eventually since it should be best for fuel efficiency and overall torque. Out of curiosity, how did you tune yours? Did you street tune it or was it dyno tuned?

Yeah my base RSX transmission's 5th gear is definitely shorter than ideal considering the torque of the K24. I think the ideal transmission for my setup would be an 02-04 RSX-S 6 speed with 04-08 TSX 5th gear, 9th gen Si 6th gear, 4.0 final drive, and OEM LSD. I'm sure the 3.4 final drive works great in an Insight, but I think it would be a bit too tall for my heavier Civic.

I think you are right and none of those are much of a concern in my application.

Alternator works great. Even with my lower idle and all electrical loads on, the voltage always stays above 14V since the Acura TSX alternator is so overkill for the small electrical demands of my Civic, especially since it has all LED lighting.

I don't have AC, so that's not a concern to me. But as weak as some AC systems are at idle, I could definitely see an even lower idle being a problem on some vehicles.

NVH is somewhat of a concern to me, but not much. It's plenty smooth at around 650 RPM and realistically a 4 cylinder on stiff mounts is never going to be as smooth as something like a straight 6 or V8, which is fine with me.

Actually at least the first of those two problems could easily be solved by simply programming the ECU to increase idle speed when the electrical load is high and/or voltage drops too low or the AC is on. As for why they don't do this, my best guess is probably because a changing idle speed could be annoying and might make people think something is poorly designed or malfunctioning.

I suspect the limiting factor is rotating mass. Compare a race-bred engine with a chopped flywheel and lots of valve overlap vs those 1930s British stationary diesels I can't recall the name of. Or a Model T Ford.

This is it in the Insight and the 3.4:

https://www.youtube.com/watch?v=h2IyVUrRaoc

I imagine you're right. Higher weight, larger tire size, 4.0 ought to be just about right.

I got an MBT timing map from Jeff Evans and did the rest of the tuning myself. It was hundreds of hours of logging. I mapped out best power by tuning all of the fuel maps to within ~1-2% for every VTC angle in every cell, then graphing them all. You can check out my K swap thread for details. It was never dyno'd.

Hmmm...there definitely is a speed below which you could accumulate some bearing wear every time the throttle is tapped to raise the engine speed. I doubt it's above 500rpm though, because engines have some lugging resistance designed in, and most engines can be run at full throttle at 1500rpm or lower. I would turn down the idle speed until one of the following happens:

1. Fuel usage stops going down despite decreased speed (unstable combustion)
2. Not enough electrical power generated to at least provide some charging for the battery
3. Engine sputters on transient load instead of quickly revving up.

Pretty much every modern car is able to idle at 650 and most "normal" 4 cylinder cars can idle at 600 if you ask them to.

Timing wise, I think the principle of operation should be "perfect is the enemy of good enough" (you can be off by a degree or two, maybe more, and not have noticeably less power), and extrapolate as much as you can based on maps you can find from other engines, and use reaction rate chemistry from high school to help you extrapolate based on load, temperature, lambda, etc.

Thanks for the responses everyone, it sounds like slightly lowering the idle shouldn't be a problem. I did notice that the coolant now seems to run slightly warmer after idling for a couple minutes, presumably due to the lower flow rate since it quickly cools off with slightly higher RPM. The difference doesn't seem anywhere near large enough to be problematic, but it's definitely something to keep an eye on if you lower your idle speed. I suspect heater performance at idle may be somewhat worse too since coolant flow seems to have been affected, but it's been too hot to test that.

I'd like to try to find a way to lower the cold idle RPM too. Even starting at 80+ degrees it initially revs to nearly 2K RPM which seems unnecessary. Since I live in Tennessee and use a block heater, it hardly ever sees cold starts below around 60 degrees. I haven't found any way to change the cold idle in Kpro, so I'm thinking of eventually installing some kind of plate between the throttle body and the IACV with a hole only large enough to allow for something like a 1200 RPM idle. Not sure how well this would work in practice, but I think it's worth a try.

What you're looking for is the "IACV duty cycle" slider. It affects the IACV for both hot and cold idle, but you can partially compensate for hot idle with ignition timing.

Thanks. I saw that, but I didn't try adjusting it yet. I didn't realize it actually changed the idle RPM setpoint. The cold idle will certainly take a lot of trial and error to get right since I don't have much time for adjustments for obvious reasons.

I don't think it actually changes the idle setpoint per se, but it holds the idle up. If you raise it, both cold and hot idle rise, and if you lower it, hot idle still aims for the target idle, while cold idle drops. Lower it too far and it will have a harder time hitting hot idle, and/or it may be easier to stall, because there's less ignition advance in reserve to control idle with.

The most efficient idle speed is 0rpm.

Agreed, but killing the engine just isn't always possible/practical.

Just loaded your 93 octane non lean burn tune and for some reason it doesn't work at all with my K24. The idle is really rough and lopey when cold, it refuses to idle when warm, doesn't want to restart when it dies out, and generally ran bad enough that I had to pull over and reload the base map. It seemed to possibly be running rich, I got a lot of white/grey smoke and it was somewhat hard to start when I reloaded the base map like it was flooded.

I'm still confused by why your tune wouldn't work for my engine since I have an almost identical setup. The only thing I can think of is maybe your fuel pressure was way different than mine. What fuel system do you have and what fuel pressure are you running?

Fuel pressure is a good bet. It's tuned for around 50psi. If you wanted to do a 5-10 minute drive with logging running and send it to me, I'd be happy to push some tweaks and we can see how it does.

I'm using a stock RSX fuel sending unit, which is supposed to produce 47-53 PSI according to a quick Google search, so the fuel pressure shouldn't be off enough to cause this much of a problem. Driving with that tune for 5-10 minutes wouldn't really be possible because it's almost impossible to keep running and might not run at all once the engine is hot. It stalled about 5 times in the maybe 1/4-1/2 mile I drove it in my neighborhood and seemed to get harder and harder to restart every time, so I didn't even try to pull out onto the highway and get it up to speed.

Maybe the problem has something to do with my engine still having the stock cam gear? Looking at the low speed cam angle tables it doesn't look like the cam should have been trying to advance beyond 25 degrees in the short time I was driving it, but it seems like that's the only other significant difference between our engine setups.

The cam gear shouldn't be a problem, it just won't use the tables if it can't advance it any farther.

I'll whip up a few changes this evening and send you a variation you can try, with a higher idle and some fueling changes.

Good to know. Since this is my first aftermarket ECU setup, I wasn't sure what would happen if the cam timing table went over 25 degrees.

Cool, thanks. I don't think the idle speed is the problem since it's set higher than what I set my base map to and it runs bad even above idle. Something with the fuel could definitely be a problem, but I'd have to get an exhaust and o2 sensor before I could do any fuel tuning. That will probably be my next project, I just need to figure out which header will work for my setup.

Edit: Does your K24 still have the thermostatically actuated air bypass valve on the intake manifold? I removed mine since I broke it shortly after buying the engine and it seems to be unnecessary from the research I did. Since that valve closes completely at somewhere around 170 degrees coolant temp and I only got up to 130-140 degrees on my brief test drive, that may be a factor during warmup if yours was tuned with that valve in place.