Go Back   EcoModder Forum > EcoModding > EcoModding Central
Register Now
 Register Now
 

Reply  Post New Thread
 
Submit Tools LinkBack Thread Tools
Old 05-18-2020, 12:14 PM   #521 (permalink)
Master EcoModder
 
Ecky's Avatar
 
Join Date: Dec 2011
Location: New Zealand
Posts: 5,077

ND Miata - '15 Mazda MX-5 Special Package
90 day: 39.72 mpg (US)

Oxygen Blue - '00 Honda Insight
90 day: 54.69 mpg (US)
Thanks: 2,904
Thanked 2,560 Times in 1,586 Posts
Over the weekend I spoke with an acquaintance who seems to have come up with a very clever economy build with K series engines, one that makes me a bit jealous. I'll do my best to explain what he has assembled in his garage, but it will require some background.

~

K series engines are very modular. The blocks are essentially all the same, and it's what's bolted to them that determines their character. Broadly speaking, they tend to fall into the categories of 1) economy vs performance head, and 2) 2.0L vs 2.4L.



Back in the 90's, Honda made a 3-stage VTEC engine, with a total of 3 different cam profiles.

In "Stage 1", the engine operates in 12v(ish) mode, with one of the intake valves opening based on the "medium" cam lobe, and the other riding on the "low" profile, barely opening (to prevent fuel pooling behind it). This causes a swirl effect which improves very low RPM combustion.

In "Stage 2", a pin locks the rockers for both intake valves together and both ride on the "medium" cam lobe.

In "Stage 3", both valves switch over to the "high" cam lobe. The exhaust valves also switch to a "high" cam lobe.

To my knowledge, only one engine was ever made that had all 3. All other Honda engines either operate in stages 1 & 2 (economy head), or 2 & 3 (performance head).


As to why you want different cam profiles in the first place, here's a chart of the fueling values for my engine, for the "low" cam and "high" cam, at various camshaft angles:





It may be hard to interpret at first glance, but essentially what happens is:

The "high" cam lets in more air at high RPM but at low RPM it produces less power, for various reasons. If the car were to run on the "high" cam at all times, it would idle poorly, have high emissions and very little power until around 4000RPM.

The "low" cam is timed to open and close more optimally for low engine speeds. Combustion quality is much better between 0-4000RPM, producing a cleaner, more economical engine with a smooth idle. However, above around 4,000RPM, the valves just don't open wide or long enough to fill the cylinder, and power starts dropping off.

Most engines without VTEC choose a single cam profile that is intermediate between these two.

My chart also includes cam phasing. Take the high cam: advance (open it sooner) by 40 degrees and it produces more power at 4000RPM. From 4250-5000RPM, it produces the most power with 30 degrees of advance. Beyond that, it should be dialed back to 20 degrees of advance. The same applies at lower speeds, but it's harder to see. 0 degrees advance creates the cleanest and smoothest idle, while different resonances produce more power at different points between 0-4000RPM with the throttle wide open.

As for part throttle, opening the intake cam sooner can allow the intake valves to be open before the exhaust valves have closed, causing some of the exhaust to be pushed into the intake manifold. This is a kind of "EGR", allowing exhaust gases a 2nd chance to let their heat do some useful work, and reducing the amount of vacuum created for a given amount of power. Too much exhaust gas mixing, however, and combustion quality drops. So, VTC can improve economy as well.

~


Between 2002 and 2011, Honda had essentially four major engine variants:

Economy K20 - found in the base model RSX and 02-05 Civic Si, 160HP peak, 12v operation, "low" compression (9.8:1)
Performance K20 - found in the RSX Type S and 06-11 Civic Si, 200HP peak, has "high" cam, high compression (11:1)
Economy K24 - found in the Accord, Element, CR-V, 160HP peak, 12v operation, has EGR valve, "low" compression (9.8:1)
Performance K24 - found in the Acura TSX, 200HP peak, has "high" cam, high compression (10.5:1)

The performance K24, from the TSX ("K24A2") is generally the most desirable engine of the lot for those looking to make big power numbers. They're so overbuilt it's possible to extract 700HP with a large turbocharger without ever opening the engine. I scored a good deal on one of these and couldn't pass it up. As it comes from the factory, it's probably the worst K series for fuel economy, with its only redeeming feature being that it's a high compression engine which makes most of its torque not too high in the rev range. To remedy this, I build an extremely tall transmission, removed the balance shafts, underdriven oil pump, lightened the flywheel, used an intake manifold meant for low-end torque rather than high end horsepower, swapped in a 50 degree VTC gear (from factory 25°), added an oil "warmer" and feed it mostly warm air. I'm also considering retrofitting the EGR system from the Accord.

~

What my acquaintance has done is very different, and in retrospect I feel it's closer to the engine I "should" have built, given my goals.

His build:
* Performance K20 block (K20A2)
* Economy K20 head (K20A3)
* Very small turbocharger (Garrett GT2554R)

Why this head?

The economy K20 head has the smallest combustion chamber, and raises compression on whatever block it's attached to. It has 12 valve operation below 2300RPM, where I operate my engine most of the time. It's only good for ~6800-7000RPM and doesn't breath all that well above 5k (hence 160HP in stock form), but the turbocharger makes up for this.

Why not just use the economy K20 engine block?

*The performance K20 block has a heavier, fully counterweighted crankshaft, and stronger rods designed to rev up to 9000RPM, It can typically take 500HP+, compared with <300HP for the economy block. Neither of these help with economy, but shouldn't harm it either, and it makes this block VERY forgiving with high compression and a turbocharger.

* Highest compression pistons of any engine block. Paired with the economy head, it has an outrageous 11.7:1 compression ratio, which should be good for ~3% economy over my engine.

* Shorter stroke than a K24. It has a more favorable rod:stroke ratio, meaning there's less side loading on the piston sleeves (lower friction), and the engine vibrates a lot less without balance shafts.

* Smaller displacement than a K24 = higher load = better economy.

* From the factory it already has an oil warmer, 50 degree VTC, no balance shafts, and underdrives the accessories (even the oil pump) compared with 2.4L engines.

Best of all, people want to get rid of these blocks and swap them for K24 blocks for the larger displacement, so they're not very expensive.

With an off-the-shelf 4.389 Honda final drive it would have only marginally higher load under normal driving conditions than my K24 with my custom (expensive) 3.4 final drive. Paired with my extra tall transmission, it would not surprise me to see 70+ mpg, so long as you stay out of boost.


Last edited by Ecky; 05-18-2020 at 01:05 PM..
  Reply With Quote
The Following 3 Users Say Thank You to Ecky For This Useful Post:
California98Civic (05-22-2020), freebeard (05-18-2020), tekcajwolley (06-02-2020)
Alt Today
Popular topics

Other popular topics in this forum...

   
Old 05-18-2020, 12:16 PM   #522 (permalink)
Master EcoModder
 
Ecky's Avatar
 
Join Date: Dec 2011
Location: New Zealand
Posts: 5,077

ND Miata - '15 Mazda MX-5 Special Package
90 day: 39.72 mpg (US)

Oxygen Blue - '00 Honda Insight
90 day: 54.69 mpg (US)
Thanks: 2,904
Thanked 2,560 Times in 1,586 Posts
Quote:
Originally Posted by California98Civic View Post
It looks great and obivously functions well or all that towing would have been a problem. Is there any way for you to track, at least roughly, it's effects on warming and cooling? Do you have any baselines you think reliable enough?
With the weather as variable as it is here, the best I can do is subjective. This month, within a 2 week period, it has 1) snowed, and 2) been in the high 70's. I think the proof will be in the averages.
  Reply With Quote
Old 05-18-2020, 04:56 PM   #523 (permalink)
Cyborg ECU
 
California98Civic's Avatar
 
Join Date: Mar 2011
Location: Coastal Southern California
Posts: 6,299

Black and Green - '98 Honda Civic DX Coupe
Team Honda
90 day: 66.42 mpg (US)

Black and Red - '00 Nashbar Custom built eBike
90 day: 3671.43 mpg (US)
Thanks: 2,373
Thanked 2,172 Times in 1,469 Posts
Gah! Snow in May! I do not miss Western Mass when I hear news like that.
__________________
See my car's mod & maintenance thread and my electric bicycle's thread for ongoing projects. I will rebuild Black and Green over decades as parts die, until it becomes a different car of roughly the same shape and color. My minimum fuel economy goal is 55 mpg while averaging posted speed limits. I generally top 60 mpg. See also my Honda manual transmission specs thread.



  Reply With Quote
Old 05-21-2020, 05:01 AM   #524 (permalink)
Master EcoModder
 
Join Date: Jan 2012
Location: United States
Posts: 1,756

spyder2 - '00 Toyota MR2 Spyder
Thanks: 104
Thanked 407 Times in 312 Posts
K20A3 head with a turbo is definitely...unusual. That's kind of like a K20C1 (probably should've just used a K20C1 unless the turbo was really cheap).

I still think you should've gone with a K20A. The fuel economy would be probably more than 10% better, no need to take out balance shafts, and you'd have 8600rpm on tap if you felt like having some fun with a Japanese ECU.
  Reply With Quote
Old 05-21-2020, 09:45 AM   #525 (permalink)
Master EcoModder
 
Ecky's Avatar
 
Join Date: Dec 2011
Location: New Zealand
Posts: 5,077

ND Miata - '15 Mazda MX-5 Special Package
90 day: 39.72 mpg (US)

Oxygen Blue - '00 Honda Insight
90 day: 54.69 mpg (US)
Thanks: 2,904
Thanked 2,560 Times in 1,586 Posts
Quote:
Originally Posted by serialk11r View Post
K20A3 head with a turbo is definitely...unusual. That's kind of like a K20C1 (probably should've just used a K20C1 unless the turbo was really cheap).

I still think you should've gone with a K20A. The fuel economy would be probably more than 10% better, no need to take out balance shafts, and you'd have 8600rpm on tap if you felt like having some fun with a Japanese ECU.
You're right, it's essentially like building a K20C.

A K20A would have been a nice choice. A few points:

I was typically finding K20A engines for $2200-2500, whereas I paid $800 for my engine. And since I have KPro, I can set my rev limiter to wherever I want. Most choose to rev the K24A up to 7600, I've limited mine to 7100 for longevity. KPro was necessary for getting all of my emissions systems working. Plus the Insight gauge cluster only reads up to 7k and an S2000 cluster would have added maybe $500 more to the total cost.

I may not have needed a $1200 final drive, but I think economy is better with the K24 and a 3.4 FD than it would be with a K20 and a 4.389.

8600rpm with my current transmission ratios is approximately:
1st gear - 53mph
2nd gear - 91mph
3rd gear - 127mph
6th gear - 265mph

Last edited by Ecky; 05-21-2020 at 12:48 PM..
  Reply With Quote
The Following User Says Thank You to Ecky For This Useful Post:
California98Civic (05-21-2020)
Old 05-22-2020, 12:25 AM   #526 (permalink)
Master EcoModder
 
Join Date: Jan 2012
Location: United States
Posts: 1,756

spyder2 - '00 Toyota MR2 Spyder
Thanks: 104
Thanked 407 Times in 312 Posts
So you're at 1950rpm at 60mph? That's pretty low. I think you would be better off with a K20 + M Factory 4.0 final drive. The specific torque output would be similar, but the K20 would be spinning a little faster, and it would do better in 6th gear at e.g. 50mph as a result.
  Reply With Quote
Old 05-22-2020, 07:54 AM   #527 (permalink)
Master EcoModder
 
Ecky's Avatar
 
Join Date: Dec 2011
Location: New Zealand
Posts: 5,077

ND Miata - '15 Mazda MX-5 Special Package
90 day: 39.72 mpg (US)

Oxygen Blue - '00 Honda Insight
90 day: 54.69 mpg (US)
Thanks: 2,904
Thanked 2,560 Times in 1,586 Posts
Quote:
Originally Posted by serialk11r View Post
So you're at 1950rpm at 60mph? That's pretty low. I think you would be better off with a K20 + M Factory 4.0 final drive. The specific torque output would be similar, but the K20 would be spinning a little faster, and it would do better in 6th gear at e.g. 50mph as a result.
I believe that's how LHT does their K-Sight.

I'd love to see how a K20 would do with a 3.4 FD, my engine is only around 20% loaded in 6th.

Thoughts on de-stoking a K24? The extra deck height would give a better rod:stroke ratio. I'd need longer rods. I think the extra dwell time at the top of the stroke improves efficiency, but it might also make it more likely to knock at very low RPM.

Last edited by Ecky; 05-22-2020 at 08:15 AM..
  Reply With Quote
Old 05-22-2020, 11:48 AM   #528 (permalink)
Master EcoModder
 
Join Date: Jan 2012
Location: United States
Posts: 1,756

spyder2 - '00 Toyota MR2 Spyder
Thanks: 104
Thanked 407 Times in 312 Posts
Quote:
Originally Posted by Ecky View Post

Thoughts on de-stoking a K24? The extra deck height would give a better rod:stroke ratio. I'd need longer rods. I think the extra dwell time at the top of the stroke improves efficiency, but it might also make it more likely to knock at very low RPM.
Destroking sounds like too much effort for the result :P You'd have a rod/stroke of 1.919 which is pretty awesome though I suppose.

I don't think very low speed knock is a problem, ignition timing is already very delayed there since there's a relative eternity for the flame front to propagate.

The two situations where I like the idea of destroking are:
1. If you have lightened the piston assembly considerably, you can increase the rod big end stiffness by reducing its diameter and destroke a teeny bit at the same time.

2. If the manufacturer uses the same size crank case and drops in different sized cranks, you can go to the bigger crank to increase stroke to make better use of the engine's mass.

I suppose if you're displacement restricted for some reason, then doing a really drastic destroke and then milling away a lot of excess material on the block is a good way to get a very undersquare engine, but it feels like a waste.

Last edited by serialk11r; 05-22-2020 at 11:58 AM..
  Reply With Quote
Old 05-23-2020, 10:17 PM   #529 (permalink)
Master EcoModder
 
Ecky's Avatar
 
Join Date: Dec 2011
Location: New Zealand
Posts: 5,077

ND Miata - '15 Mazda MX-5 Special Package
90 day: 39.72 mpg (US)

Oxygen Blue - '00 Honda Insight
90 day: 54.69 mpg (US)
Thanks: 2,904
Thanked 2,560 Times in 1,586 Posts
Moved 2087lbs today, averaged 37mpg. It felt more like a "normal" car.




I didn't expect it to be half that heavy, and was surprised when I did the math at home.
  Reply With Quote
The Following User Says Thank You to Ecky For This Useful Post:
California98Civic (05-25-2020)
Old 05-25-2020, 10:51 AM   #530 (permalink)
EcoModding Lurker
 
Join Date: Jul 2019
Location: Washington State
Posts: 15
Thanks: 5
Thanked 4 Times in 4 Posts
That's pretty cool that you can tow a ton. I flat-towed my brother's VW rabbit to a gas station once with my 90 Integra (on the way to a VW meet where they liked to pick on Hondas a bit lol), and it was not a pleasant experience - definitely didn't have enough power for that.

I've always wanted to build a small displacement, longish stroke, high compression economy/mild performance engine with a smallish turbo. In theory that should be most efficient. I forgot you were running a 2.4l; I'd love to see what your hypothetical K20 build would do for economy, and with the right turbo you can essentially pick how much power you want (within reason and the limitations of your head flow, block strength, intake/exhaust, drivetrain strength, etc.)

I would really love to do an Insight based build one of these days. Sounds like a fun project but I doubt I'll ever have the time to make it happen.

  Reply With Quote
Reply  Post New Thread






Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2024, vBulletin Solutions Inc.
Content Relevant URLs by vBSEO 3.5.2
All content copyright EcoModder.com