D15B2 All-Motor build

rushi · 06-14-2009, 05:53 AM

So many have requested me to do a build thread about my current D15B2 build, and I've always told that yea I'll do it when I've time.. I decided that today, a dull Sunday noon I've time for it so here it comes!

You probably know that I've been planning to make my already powerful D16Z6 even more crazier. It takes long time and a bunch of money too, engine is disassembled and lacking parts so I needed an engine for my car so I can drive it. This is where this build started. My original plan was just to mock this engine together with used parts and minimum cost. I slipped from the original plan first time when I measured the piston/cylinder clearances. They were totally worn out, far from service limit and I know from experience what would happen if I'd assemble the engine like this.

So, I bought some new 76mm YCP PM3 pistons from FJT among with gaskets etc. Now I had decided that I'll make this a proper rebuild with mostly new parts, to make it good and reliable backup engine. I had already bought my another '89 D14A1 hatch and needed to get the car running ASAP as I got off the Army and started to work, so I just pulled the D14 with tranny and put it into my ED. More about this engine and setup can be found from this thread.

This is pretty much the point when this "project" got totally out of control. I got an idea that maybe I should try to make this engine as efficient as I can, trying to get maximum fuel efficiency out of it. Another usage would be to use it as experimental testbed for new ideas that I wasn't sure do they work or not. I'll show here what I've done to various parts in my engine. Actually, there is not many parts that aren't modified somehow

I won't show or tell everything, but if you've sharp eye you might notice some unmentioned things in following pictures..

I'll start with the foundation, engine block. I had already decided to bore it to 76mm to get maximum displacement, piston area and deshrouding. One thing I really wanted to try was block posting. I bought 8mm aluminum rod and threaded it to M8 thread to form the posts itself. Next I had to determine the proper placement of the posts in block. At the power stroke piston starts to descend and because of the engine rotation, rod will turn towards the exhaust side of the engine. This rod angularity combined with the force (combustion pressure) applied to piston causes the highest cylinder wall loading towards the intake side of engine. I thought that there might be two points where the load is highest; At the maximum cylinder pressure and at the biggest rod angle. I didn't calculate this anyway, just a though, but I decided that is close enough to reality. First case, maximum pressure, should occur at around 12-14deg ATDC according to many books. I roughly measured how much this is in millimeters and added piston compression height to that amount (highest piston side load is at the piston pin height). This is the distance from deck level for the upper block posts. Another one is easy, biggest rod angle is of course at the half stroke. Again, half stroke plus CH, 84.5mm/2 + 30.7mm is the distance from deck for lower posts.

As the loads on exhaust side are not that great (no much pressure pushing the piston against wall) I though that one post per cyl is enough, and I decided to place them roughly in the middle of the intake side posts. I secured them in place with Devcon F Aluminum Putty (thank you Luke for it!), and I lightly etched the posts in mild hydrochloric acid solution.

Next thing in block is minimizing the parasitic losses and drag. According to Endyn's D-series tips, I sharpened the crank girdle edge so it'd shave the oil off the crank a bit more efficient.

Next is crank and bearings. I used ACL Aluglide main bearings and NDC rod bearings. Again according to Endyns tips, I enlarged the main bearing oil holes. I did this in drill press which I don't recommend. I'll tell you later that I had to disassemble the engine soon I got it running, and I noticed some crank to bearing contact around the oil holes because of bearing deformation..

One thing I like in D15 is that it has about 4lbs lighter crank than D16. It also has four counterweights slightly smaller than other four, which is a disadvantage but I think we can live with this. Both main and rod journals are also smaller in diameter (mains 45mm vs. 55mm, rods 42mm vs. 45mm) which result smaller parasitic drag. Crank itself is forged like all Honda cranks I've seen so it should be reasonably durable for naturally aspirated engines.

Crank got some extensive modifications. First was oiling, I chamfered all oiling holes and "fluted" them at main journals so oil would turn easier into the passages. I entirely polished the whole crank with first 60grit, then 120grit flapper wheels. I dunno which kind of effect this has for crank durability, but I'll find it out. Like I said this is experimental engine

One more crank mod was to "aerofoil" the counterweights. Rounded leading edges, tapered trailing edges. I did this work with angle grinder, and surprisingly after all these mods the crank was in perfect balance!

Mated to the crank is Fidanza flywheel that I originally bought for my Z6. As I wanted the least possible inertia, I decided to drill 33 15.5mm holes to it to reduce some weight. This ended up from stock 3120gram to 2860grams. Clutch is unsprung Action 1MD from Bisimoto, again for least inertia and fastest shifting.

At the other side of crank is modified new OEM D16Y8 harmonic damper. Since I'm planning to spin this thing to 9000rpm, I want to ensure the crank and oilpump durability. This damper is huge and in stock condition it weights about 4 kilograms! We turned the power steering part off to make some clearance and remove all unnecessary weight. I think this should balance the rotating assembly quite nice, since it weights almost the same as flywheel+clutch combo! Of course the inertia is lesser because of smaller diameter..

OK, I'll take a break now, but I'll continue soon! Most likely later today.
I've a small request according to this thread.. Please keep the conversation (that is very welcome of course!) strictly technical. I don't like it when threads like this get full of bullshit and grow into 20-page monsters that noone wants to read and finding particular info gets very difficult.

rushi · 06-14-2009, 08:46 AM

Oilpump is modified D16Y8 unit. Why I choose this pump is primarily the OEM cranktrigger setup that can be used with VEMS standalone. Crank trigger is VERY important for precise electronic engine control. Timing belt itself can cause timing variance of several degrees, and add to this the camshaft twisting and springing, and you'll quickly understand why trigger sensors mounted to distributor are totally unacceptable to any performance engine. I believe this is one reason why newer K- and F-series engines are so superior to older B/H-series. You know how important it is to tune the timing map precisely at high rpm.. It might be a matter of a couple of degrees between the maximum torque and exploding the engine. Now if your timing is jumping all around +/-3...5 degrees at 9000... Needless to say more. Of course crank itself is twisting too, and optimal trigger placement would be in the middle of the crank. This is impractical, and next best would be the flywheels since there the engine position is referenced to output. But in reality the timing end is very good too. I removed one tooth from the pulley, so the edge after "missing tooth" occurs at 60deg BTDC. This makes it possible to use wasted spark ignition with VEMS. As you probably know I'm running COP and sequential injection which requires also a cam trigger, but this is still a nice option. More about the COP/camsync in my D14 thread.

Many have questioned the Y8's pumping capability and several are directly blaiming this pump for their engine failures. It sure has its faults, but those can be corrected quite easily like I've described in my oilpump modification thread. I actually see this probably reduced pumping volume as a benefit. Its needless to pump more oil than is needed since it only robs the power from crank.

Oil pan is standard D16Y8 cast aluminum piece. I like aluminum more since it allows the oil cool better in hard use. It also mates directly to Y8 pump which has slightly different stud pattern than older pumps. Y8 pan has also removable steel baffling which makes it easy to clean and modify. Thank you Remoer for the oilpan! Only modification I did to the pan was additional hole for dipstick since I'm using standard front-mounted dipstick in D15 block. If I'd originally knew that I'm also gonna use Y8 pan and timing belt cover with the Y8 pump, I wouldn't plug the dipstick hole in pump. I'd actually plug it in block and use the oilpump-mounted dipstick. It looks more modern and makes room for wild header configurations

Other considerations about the lubrication system are the filter and oil. I use Honda OEM S2000 filter, 15400-PCX-004. Oil I'm using is Motul 300V Motorsport 0W-30. It is light oil for very powerful and high revving engines, its based on 100% polyolesther (similar like Redline) and it should be one of the best oils available for normal mortals. Four liters of this stuff cost about $100...

Talking about pumps, lets take a look at water pump. Biggest modification to new OEM part here is the 22-tooth B16 WP pulley. Stock D-series one is 19-tooth. This serves two purposes; to reduce the pump RPM and therefor cavitation and drag at high RPM, and compensate the .110" head mill. Timing belt is stock D15B2 Gates 106-tooth belt. My head milling thread has some info regarding these parts.

Next part is the connecting rod. I said before that I used NDC Bearings from FJT. Rods are pretty much stock D15 pieces. They have ARP bolts, and necessary big end honing/reconditioning all done by machine shop. According to few persons, some are shifting at 9500 with totally stock rods and the beam strength shouldn't be an issue with these so I'm not worried about them with the improved oiling system of my engine.

Pistons are tricky. I originally got those 76mm YCP PM3s from FJT that got extensively modified. I originally calculated that I should have 13:1 compression with my .11" milled head assuming zero cc flat top pistons. This of course is not true because pistons have valve reliefs on them. I measured the stock pistons reliefs being 1.5cc, but because of the extreme head mill I had to widen the reliefs so much that they ended up being 4.5cc! This with pistons coming .005 over the deck and .045 Cometic head gasket yielded 11.6:1 static compression ratio and I got really disappointed. This is one point of learning, and gaining experience is one of the primary goals with this engine.

I also drilled eight 1mm gas ports to the pistons, and ring package I used was constructed from two different ring sets to get least possible static ring tension and friction. Top compression rings were molybdenum type instead of typical Honda chromed rings. Molybdenum is porous (holds oil) and has lesser coefficient of friction that chrome, so this was also one trick to better mechanical friction.

Why I'm talking in past form with these parts is that I actually got the engine running, and drove it about 200 miles. Then I blew the ringlands of #1 piston because of detonation at 6000rpm. There is several reasons for this. One for sure was excessive ignition advance. I was running 33deg at 6k, which is far too much for such of a fast-burning engine. Gasports might have their share of the damage, since they let the detonation pressure to hammer the ringlands directly. Its sure indication of detonation that it was the exhaust side ringlands that got blown (remember piston is pressed against the intake side cylinder wall during the power stroke) exposing even bigger gap to exhaust side. There was also a noticeable contact marks in all upper rod bearing shells at the corresponding rod angle that prove it was detonating in all cylinders. Another reason for detonation and especially #1 damage might be the #1 running slightly lean with my modified Blox IM. #1 runner is very close to plenum end wall and pressure might be packing there. I made a notice when I disassembled my D14 that #1 plug was slightly lighter color than others.

So the current situation is that engine is assembled again (I had to re-hone the cylinders lightly, used 400grit Flex-Hone) and I'm waiting new pistons from FJT. This time I'll try 76mm Nippon PG6 pistons with Nippon chrome rings. PG6 has very small dome and very small valve reliefs, so I can enlarge the reliefs just the amount I need (PM3 had too big reliefs; hell I could push the valves 6...8mm open at the TDC!). With these pistons and big-enough valve reliefs I'm expecting to see 12.5-13:1 static CR. I just gotta be careful with the ignition timing..

This time I'll also leave the gasports off. They might work well, but I think they make cast pistons too fragile and sensitive to detonation. The moly rings also got damaged from the detonation (moly inlay got slightly chipped off) so I'll go with Nippon rings now. I don't like Hastings rings since their top ring is super-stiff.

To be continued. Another break, and next I'll show a little something about the cylinder head..

Sever · 06-14-2009, 09:23 AM

couple of quick questions.

how much did your crank end up weighing?

and were there any balance issues after drilling that many holes into the flywheel?

rushi · 06-14-2009, 02:24 PM

I didn't weight the crank. My intention was not to lighten it, just make it more aerodynamic. I think it lost less than one pound.

No balance issues. Holes in flywheel are symmetric. Crank was in perfect balance, for flywheel they had to drill like two 5mm holes few millimeters deep. Thats almost nothing.

OK the next area is cylinder head. This is where the most work is done with great results. Basicly the head I started with is PM3-16 casting from carburated '89 D14A1. Since its carb head, it doesn't have EGR passage drilled like FI heads have and therefore all exhaust ports are identical. It has opening for mechanical fuel pump, but that is easily plugged with a plate I made from 5mm aluminum.

I pretty much used all new parts for the head. SI Valves manganeze bronze valve guides and stainless valves. Intake valves are +1mm oversize (30mm), exhaust valves are std size 25mm. This is because I wanted to maximize the intake flow, and with high compression the high exhaust flow isn't needed. Well this ended up pretty much opposite, more on that later. About those SI-valves, they're horrible quality and needed lots of work before installation. seat width on intake valves were like 2...3mm wide and I had to do extensive 30deg backcut into all valves to get seat width down to 0.8mm intake and 1.2mm exhaust. Also the exhaust valve stems were really badly undercut, there was sharp groove in every stem and I had to extend the undercut and polish the stems. Its also weird that undercut is very short, only to halfway towards the guide. There's clearly a reason why these valves cost only $189 per set. I won't recommend them to anyone nor will I use them myself again. Other valvetrain parts involve Bisimoto Pro spring and Ti-retainers. Those are nice parts with great fit, highly recommended!

I did lots of work with the combustion chambers. I already told before that head is milled .110" (2.7mm). This is because I wanted the chambers as compact as possible to get tight squish over a large area. Lots of issues rose from the extreme mill and those are covered in the head milling thread.

Next pictures are before-after from completely stock casting to my finished piece:

Porting is actually quite straightforward, especially on exhaust side. All I did was clean up the bowls, smooth the guide/boss into roof and sharpened the divider. This together with radius valve job made Serdi resulted in 175CFM at 28"H2O and 9...12mm lift. According to the flowbench(SF-600)/tuning shop operator this is enough for 220-230bhp engine and excellent flow considering the tiny valve size.

Intake side is a bit more tricky. I first tried the same basic work like in exhaust ports together with port slightly widened and gasket-matched to Z6 IM gasket. Again Serdi radius valve job was used. Result very disappointing 180CFM at 28"/9-12mm lift. Operator told this is only enough for about 190bhp engine. He told that he tried to add a couple of millimeters of BluTack to short side turn, and flow immediately improved a lot. So I got the head back to home, roughed the port floor and added about 5mm of Devcon into it. Now the port CSA at the "throat" is maybe 1/3 smaller than before, for sure the velocity is equally higher and the peak flow increase was dramatic. The whole intake flow curve is raised few CFM from the point where valve just cracks open all the way to peak flow that ended up being 193CFM at 11mm lift. What is interesting is that the biggest flow gains happened between 9 and 11mm lift. It flattened out at 9mm first, now it keeps climbing upto 11mm. This should be enough for 215bhp, and that would be nice to achieve

I do not have any pictures online from the ports, sorry. Actually I'm not much willing to share the pics publicly especially from the intake ports and I hope you understand this. A couple of shots from other valvetrain parts and valvejob:

I need to resize and upload a couple of pictures from reworked valves, and maybe something from the porting.. I don't have Excel in my computer atm so I can't make nice graphs from the flowcharts now. Let me know if you're interested about them, I can make those later.

I have basicly three camshaft options right now. First is stock D16 cam that I only intend to use for inspection/emissions check. Next is Bisimoto level 2.2 cam. I bought this cam used and it was made for D16/ZC. It didn't work very well with D14 (not much improvement from stock D16 cam), I'll see how it performs with this engine. Last one is the most interesting Exospeed Stage 4 HC cam. I really liked this on D14 and I'm really excited to see how it performs with this D15. Rocker angles end up quite bad with this big regrind and some ppl say it'll kill the valvetrain/guides very fast. Exospeed/Wil says they've made over 50 of these and never had any issues, tens of them being used on streets/DD with no longevity issues. I'll run this and see does anything wear/broke out. Again this is experimental engine. This cam is mated to Bisimoto cam gear.

Here is a pic of how it compares to stock D16 cam. Pretty HC huh;

What we've left is valve cover and breather setup, manifolding and rest of intake/exhaust systems. VC/breather is easy, I haven't done them yet

But I'll make a couple of vents into the VC and block (modify the "black box") and route them into separate breather/catch can similar to Endyn/Moroso and T1 style with a filter or few.

Header is still the same 4-1 venturi merge anti-reversion piece I made years ago for my Z6. This header has worked excellent since the day I made it. I just noticed its starting to rust excessively under the heatwrap (it even feels soft and crunchy!) and I've been planning to make a few new headers with different designs and dimensions this summer. Exhaust is still the same 2.5", might redo it too. Probably try 3".

As for intake manifold I now run my modified Blox IM with RDX injectors. Another manifold I'm experimenting is described in TIG-training thread. Still unsure about the injectors for this setup. I might try stock ones, get slightly bigger RC injectors or then adapt RDX ones (the best option). I'm also working on 44mm ITB setup with eight injectors staged.

I'll also experiment with different intake tube designs and setups. Most interesting is the one we've been working with few guys here (Hondanickx and Glenniz) utilizing the aluminum radius adaptor based on TOO's R=1/2D idea.

Well this is it this time. I'll add information, pics and videos as I proceed. I told I had it running for 200 miles before it blew, and I can tell you that it ran super-well and strong. I had Bisi 2.2 cam in. It accelerated/revved like a motorcycle, very rapid. And while driving it accelerated hard with just blipping the throttle. At city speeds and with lower gears, it break the tires loose just by kicking the pedal in. It was real blast to drive! You can just imagine how excited I am to get this back on the road ASAP!

I hope you my friends enjoy this. I used almost 4 hours for writing all this crap. Thread is open for all kind of tech talk, please chime in!

ngata_speed · 06-14-2009, 09:54 PM

Wow Rushi, really nice build. Lots of hard work going into it. Thanks for posting this up, very informative and good luck with the build.

SOHCinWA · 06-14-2009, 10:30 PM

Thank you rushi for sharing all of your hard work with everyone.

BoneSpec · 06-15-2009, 01:25 AM

giggling like a teenager.

Esp. the flywheel and crank pulley setup....as long as you have enough to launch it.

Remove weight the farthest from centerline while keeping the heavy damper, perfect!

IMHO still too light for drag racing though

rushi · 06-15-2009, 02:39 AM

Heh glad you like it. Don, you can be sure I'm not showing "all" of my hard work.. What I showed is just a scratch..

Bone yeah

But don't worry, its not for dragracing

although I'm planning to try it next month..

sl@sh · 06-15-2009, 07:02 AM

good work rushi! keep it up...

this should answer most of the questions by others to rushi..

but then again showing innovative work.. tend to add more questions from others haha

either way.. GOOD STUFF!

FURYOF4 · 06-15-2009, 08:29 AM

Great work as always Rushi! Thank you for sharing.