Hyperblade's KP61R - Toyota Starlet with Honda K20a

[Hyperblade]

So what everyone has actually being waiting for, the final comparison between Toyota Altezza RS200 3SGE BEAMS and a Honda Accord Euro R K20a engine in the same chassis

Still trying to get my hands on a single dyno sheet with the 2 different engines on it. So this will have to do.

Modifications/Setup

Toyota Altezza RS200 3SGE:

• 120,000 k's
• SQ Engineering Quad Throttle adapter plate
• 4AGE Blacktop throttles
• 70mm trumpets
• Combined Piper Cross Filter
• Stock Injectors
• SQ Engineering Slim line alternator (smaller Echo alternator)
• No Power Steering pump or AC
• Custom header (TRD Copy) to side exit exhaust 2.5" Adrenaline R muffler
• Link G4+ Storm (Blue)
• Engine internally stock
• Lightened flywheel
• HD Exedy Clutch
• J160 Gearbox
• 3" Driveshaft with sliding yoke into gearbox

Oil: Castrol 5w40 
Rev Limit: 8000rpm
Dynoed in 5th gear 1:1.

Honda Accord Euro R K20a:

• 108,000 k's
• Skunk 2 Ultra Street Intake Manifold
• 770cc FIC injectors
• Skunk 2 74mm Throttle Body.
• 3" intake pipe from pod filter.
• Custom header to side exit exhaust 2.5" Adrenaline R muffler
• Stock water pump replaced with EWP
• Alternator replaced with smaller Honda D15 one.
• No Power Steering pump or AC
• Stock Oil Pump and Balance Shafts replaced with ported Type S oil pump.
• New Timing chain
• Lightened flywheel
• HD Exedy Clutch
• Link G4X XtremeX
• Honda S2000 AP1 Gearbox
• 3" Driveshaft with sliding yoke built into shaft as gearbox has flange
• VTEC 4500rpm

Oil: Castrol 5w40 
Rev Limit: 9000rpm
Dynoed in 5th Gear 1: 0.94076

Same:

• Tires 54cm Michelin Slicks at 21 PSI, on same Rims
• Same AE86 Diff (Brakes were changed from stock to AP with knock back) 4.5555 Crown wheel and Pinion.
• Same Dyno

Different:
Dyno retarders have changed from 110v to 220v.

 

Results
3SGE :
201hp @7250rpm
173.6 lbft @5150rpm

K20a:
206HP @8250rpm
167.9lbft @5300rpm

Thoughts
I'm a little bit disappointed in final figure, but I think my expectations were to high.

You have to take all the HP figures that everyone posts with a grain of salt. 
Especially anything from the UK who like to give made up flywheel figures when they dyno there cars on a rolling road...

Changing from a front wheel drive to rear wheel drive setup definitely has more drivetrain loss, from what I've read a factory stock S2000 AP1 with 240/250hp at the flywheel dynos 200hp at the wheels.

Also a lot of people show "stock internals" figures which include changing cams etc.

So the power figure is in the right ballpark really for the modifications (stock K20a has 220ps at flywheel) as rear wheel drive.

Looking at some of the NZ figures I have seen, the engine would probably get quite a gain (20 to 30hp) from a set of drop in cams (no other valve train changes)

But if you ignore the single figure and look at the dyno sheet it pretty much makes 200hp from 6700 rpm all the way to 8700 rpm which is a very significant power band.

So overall, it's a good improvement.

 

Shakedown
So I then got it out to the track and managed 4 laps before the gearbox lost most of it's oil when a bolt came loose. Luckily I was going fairly slowly so no internal damage done hopefully.

However a bigger issue is a massive driveline vibration.

Plan A: Driveshaft is getting balanced at 5000 rpm (it does 9500 rpm in 6th)
Plan B: CV Front joint on the same shaft, or a replacement driveshaft.

Driveshaft is to short to go 2 piece and we think the angle of the engine and gearbox is impacting the universal joints. However aftermarket CV joints for these are supposedly not very good. So currently looking at options.

Car definitely felt like it was pulling really hard, but with so few laps, and me just making sure everything was working, not a good indication of final result.

 

Videos:

First 2 Laps:

 

 

Second 2 Laps (where it dropped the oil):

Flyby (Potato Cam):

 

 

[Hyperblade]

Some progress has been made.

Driveshaft was sent up north to be balanced at a higher speed and the UJ's were tightened.

And so had it out for another shakedown at a local sprint day.

First race just took it easy checking things over (pretty boring video, best to skip).

 

 

Second race, gave it full beans off the line against an EG civic with K24 and 300hp at the wheels, he got a bad start...

 

 

Third race on slicks and it started drizzling, but overtook a Porsche and a Lotus...

 

 

The driveshaft was a lot better, enough for me to push it a lot harder in those videos.
But there was still a vibration there that won't be good for components, best guess is it's the angle of the engine/gearbox in the car, front of engine angled to left to clear headers and for weight distribution which means the UJ's are not in sync.

Those videos also have me mucking around a lot with the brake bias as that's been all reset, along with suspension changes affecting rear of the car during braking so still getting it dialed in (sprint day you don't get many laps).

Other thoughts, is the S2000 Gearbox throw is as short as the J160 with a short shift kit, if not shorter, it's a great feeling box.

Also you may have noticed in my older videos I didn't heel and toe, originally that was because it was impossible to with the stock pedal setup and my seating position, 
but since I have the new pedal box and seat I really needed to learn, so with the minimal seat time I've been trying to get used to it, still a work in progress as you can see in the videos.

So the shake down highlighted a few issues.

• Driveline Vibration still there
• What we believe is heat soak of the oil temp sensor as it hit 123 degrees, and there should be no way the oil hits that when it struggled to get over 80 on dyno (with some of my bonnet ducting removed which allowed a breeze over the sensor).
• Getting car up to temp with the EWP is a pain, it's not quite right at the moment. Was setup on dyno perfectly, but the real world has a lot more airflow over radiator.

So before Christmas I shopped around for a driveshaft with a CV on the front, basically came down to only one place who were prepared/could make it to handle 9500rpm.

Hence a new Carbon fibre driveshaft arrived last week from the Drive Shaft Shop out of the US, i'll be honest I was very hesitant in ordering from them as I had been forewarned by a couple of people about their quality and after sales support.
But I had no other options. I can say their pre sales service was good with prompt communication and fast building of it and shipping. 

It still needs to be fitted into the car, we are just making up a flange that goes over the existing AE86 Diff flange (from the diff side) to better spread the load from the new driveshafts rear Hilux UJ which has a bigger bolt pattern.

Here's the difference between Hilux and AE86 diff flanges.

Heat soak of the oil temp sensor should hopefully be solved by some new shield i created.

The EWP control is a bit if a pain, the duty cycle is controlled by RPM vs ECT. 
Which is great but i feel it misses another key input which is the speed of the car which affects the car's ability to get rid of heat.

So it feels like half the solution to the problem.

You can see the duty cycle all over the place here and the pump is barely breaking a sweat (and that's with half my radiator ducting closed off).

For the EWP I made up (by hand, no lathe...) a reducer to drop the coolant pipe size down from 30mm to 18mm to slow the water, it goes into the return feed on the radiator after the water goes through the engine.

I'm hoping that means I can use more duty cycle in the pump, so give a wider duty cycle range to play with.

I've then tweaked the table handling the ewp to try and get it to be more flexible in it's ramp up/down by letting the ECU interpolate in between the cells more, feel free to let me know if this is a stupid idea...

Old: 

New:

So once diff flange is done, will aim to get it back out for another shakedown to see if we have finally solved the vibration and see how we are going with ECT.

 

Discussion here: 

 

[Hyperblade]

Flange has been done, and will get fitted this week.

I can run the current driveshaft I have or the new one. Just press fits over existing flange.

Original flange probably would have been fine with the forces involved and the fact it would have mainly been a clamping force. But I decided it's just not worth running the risk considering the catastrophic consequences that could occur.

Come up great.

So it turns out 0.5mm aluminum wont withstand 200km+ who would have guessed, I am surprised it actually survived a few laps.

I had strengthened the bottom edge, but ran out of material before I could do the rest and got the car on track.

Repaired version, with nice rounded edges for the airflow (not that I need any extra, keeping the car in the temperature zone is actually the biggest issue).

Painted and it fits in nicely (ignore raw unpainted panel below, at some point will get splitter off and paint that, but not a priority).

Sharp eyes will notice the grill is open, once car is fully shaken down and working how i want, then I will blank that off and feed cold air into the intake from the left headlight.

Very long term It would be nice to have a fibreglass/carbon piece that replaces the grill, headlights and side lights as one piece, then put a vinyl photo on it to make it look normal. would save a couple of kg and be a lot more aerodynamic.

[Hyperblade]

On 09/02/2022 at 20:17, Hyperblade said:

So before Christmas I shopped around for a driveshaft with a CV on the front, basically came down to only one place who were prepared/could make it to handle 9500rpm.

Hence a new Carbon fibre driveshaft arrived last week from the Drive Shaft Shop out of the US, i'll be honest I was very hesitant in ordering from them as I had been forewarned by a couple of people about their quality and after sales support.
But I had no other options. I can say their pre sales service was good with prompt communication and fast building of it and shipping. 

 

So...

Diff flange was all fitted up and the driveshaft was fitted and then checked against the instructions from The Driveshaft Shop, and it was too short...

It was extended as far as it could go so could not thrust out if required.

To say I was disappointed would be an understatement. I had even double checked the measurements on the quote via email before ordering and asked if it took into account the thrusting and they said yes...

When building a driveshaft it's a fundamental measurement, to fuck that up time and time again as I had been pre warned shows that they don't have the process and checks and balances in place to ensure the consistent quality of their products.

 

I contacted the original salesman and he was very responsive, he got me to take measurements with driveshaft in place and quickly had a spacer for the rear diff made and shipped out to me at their cost. (luckily the driveshaft was short, if it was to long that would be an interesting discussion...)

 

Once I had the spacer I made up the heat shielding for the front of the driveshaft, the carbon shaft has warnings on how hot it is allowed to get, and also I want to protect the CV joint from getting exposed to to much heat to prolong it's life.

 

I have a new speed sensor on the way as my one is behaving really weird (cutting off the signal at 5000hz), hoping that solves the problem.

Next step is get the car out on track again for a shakedown to see if I've improved the way the cooling works and fingers crossed the driveshaft solves the vibration issues.
 

 

 

[Hyperblade]

So had the car out to a track day with the new driveshaft and I can finally say the vibration issue is solved.

So that's a massive win.

Here's some footage (track day so no passing allowed on corners)

 

 

 

Car Handling 
Car handling is taking a bit of getting used to, traditionally when i was mid corner and I put the power down it would want to oversteer. When you get on the power now it actually get's more grip in the rear but also turns in more, which is a very unusual feeling to get used to. But a positive one that I need to adapt to.

I ran a 1:36.500 without going crazy on 4 year old tyres and 15 degree day (with only one clear lap), which is down to what I was doing before, so that's nice to finally get back to that. Next target is PB of 1:35.800.

 

Water temperature

Water temperature is still an issue however, just hard to get it up to temp quickly. Which won't work for one warm up lap then standing start race.

I've been trying to avoid putting in a thermostat as there isn't a lot of room and it's a clean simple setup at the moment (the stock setup is long gone as it weighed 5kg and was in the way of the engine mount)

So I finally bit the bullet and after 2 weeks of mucking around with stores in the UK finally managed to convince someone to take my money and ship one out to me.

I got the smallest I could find, it takes a classic mini thermostat (82c). I've drilled a 2.5mm hole in the thermostat to keep water circulating through at a reasonable speed.

 

 

Space is very tight to fit it in and I didn't want to muck around putting it under the intake manifold as it would make it a real pain to service if needed.

I only just managed to fit it at the top inlet on the radiator with some gentle persuasion of the oil cooler ducting, not pretty but it works.

 

 

EWP Setup
So my original EWP setup (by the tuner) was a running the water pump at Frequency of 0.6Hz (we will come back to this) and I had been changing the table to try and get it to find a happy medium by itself, so between 80 and 92 degrees letting it scale up as required. 

 

This just wasn't working (without thermostat), and I was concerned with the frequency what impact that would have on the pump life as it was basically stopping and starting it 0.6 times a second, which didn't seem like a good idea (but I'm not an expert in this area).

So I sent support email to Davies Craig (who make the water pump) for advise on its limits as I was a little concerned by how it had been setup by the tuning shop.

Their support staff rung me and spent over 45 mins on the phone to me going through the possible causes and solutions. That's amazing service for just a simple water pump.

They said the pump should be fine for my setup (I was thinking it was overkill) and that they hated restrictors and didn't recommend them due to it restricting the max flow of the pump.

The best guess was that the radiator was super cooling the water, so when running through the engine it was to cold and so never allowed the engine to get to the right temp, and then when flow increased it dumped a lot of cold water into the engine block, detected it was cold then slowed the pump again.

Basically the pump was running too slowly (which seems exactly like what you want when your trying to get the car up to temp).

They also confirmed running it at 0.6hz was a bad idea, after some discussion on the pros and cons of higher vs lower frequency (higher = less noise on the circuit) we went with 100hz.

And a completely different table.

 

The idea is the thermostat always has steady stream of water flowing over it through the 2.5mm hole so it's bulb will always be reacting fast to the water temp, it allows the whole circuit to heat up and also means the water is flowing through the block faster so means more consistent readings at the ECT sensor and spends less time in the radiator.

So up to 82 degrees the thermostat controls that flow, as soon as it goes over that the thermostat starts opening and allowing more flow so so should self manage it, the EWP is then set to max out it's flow as we get closer to 100 degrees, so should eventually settle at a happy medium.

Testing on jack stands shows quite a different characteristics in how it gets up to temp and holds it, so I'm cautiously optimistic we might have solved the issue (bar minor tweaking).

But on track is vastly different in terms of air flow, so need to get it out for another shakedown.

Racing starts 14th August, so need it all sorted by then.
 

[Hyperblade]

So back out for a shake down with the new thermostat and when the car was allowed to sit in pits for 20 mins it could get up to temp can maintain it out on the track when pushing hard, however as soon as I eased back it dropped to 70c then had trouble getting backup to 80c.

I did manage a new PB though of a 1:35.100 which i was pretty stoked with.

 

So getting completely over the whole temperature issue by now and the first race day coming up I decided to put a bypass hose in from the thermostat back to the pipe going into the water pump.

 

Idea being the thermostat only lets in cold water when it needs to.

So while testing that on jackstands, I happened to have my dad over who was checking out the temps on the hoses as it warmed up, and he noticed the thermostat appeared to open at 70c, which for an 82c thermostat shouldn't be possible (as thermostats these days are fairly reliable in their operation).

Some head scratching and thinking later...

In the link firmware I had originally configured the base map off one LinkEcu had done, and they had manually added in K20 temp sensor calibration.

 

Since then LinkEcu have released a plugin ECU for Civics with K20's and they have set up a pre selectable calibration for "Honda K20" temp sensor.

 

Swapping over to that one immediately increased the ECT temp by 9 degrees, meaning it's been running a lot hotter then I was expecting and throwing all the water pump stuff way out.

Having seen numbers of 92 degrees in my logs which in reality meant 101 degrees, that's a little bit concerning, we don't think anything's been damaged but it could have been a lot worse...

Once that was changed, and with the new bypass the car warmed up within 5 mins, and was a lot more consistent in it's operating temps.

 

Also found my rear brake pads were not the compound we all thought they were (they were more aggressive then the front), so put a new less aggressive set in

 

So going straight into race day and into Qualifying my goal was to make sure Water temps were ok and bedding in the rear brake pads.

Qualifying went well with a 1:36.800 which was pretty good considering I was busy doing the above.

Water temp was consistently over 80 degrees even when just driving easily around the track.

The new brake pads were a big improvement, I had previously been struggling with inconsistent rear lock up, it turns out the old pads would heat up then suddenly aggressively bite which immediately wants to make the car swap ends, to combat that I had been braking lighter and longer which was affecting hitting the apex and general braking distance and also my confidence in passing under brakes.

 

So lined up for Race 1 and got a brilliant start, was doing well holding off a high powered 4wd Subaru, then on lap 2 my clutch pedal went to the floor and wouldn't come back.

 

Looking underneath the car we can see a clutch plate spring coming through the release fork hole in the gearbox...

It was an S2000 Exedy HD clutch in a light weight car which hasn't done a lot of work, so pretty pissed off about it all (the same type with the 3SGE lasted over 4 years of hard racing)

So currently looking at options for replacement, and hoping there isn't to much damage in the box.

[Hyperblade]

Clutch:

So the aftermath of a clutch spring letting go.

 

Bit's also made a small hole through the bellhousing.

That was a new Exedy HD S2000 clutch, there were warnings on the internet about using them, I thought it would be ok as light weight car, and previous Exedy with the BEAMS handled a ton of abuse.

Wow was I wrong, DO NOT use this clutch, after taking it out we could see wear on the posts and clutch springs where they were binding, it's just not designed correctly for the loads.

Decided to try an Action Clutch (Considered Xtreme Clutch, but had a friend who's plate exploded on his car and made a bigger mess...) this time, at least the springs shouldn't be able to come out...

 

Adapter Plate/Flywheel

So while we were swapping that we found the input shaft bush was showing a lot of wear and was no longer concentric. 

I run a standard adapter plate (with flywheel to suit), they say on their product page "Precision machined from 6061 this adapter features perfect dowel pinned alignment and all hardware needed."

Of course when I went to bolt it on one of the dowel pins wouldn't go in, I was in a rush for various reason and it become a one is in, that should be enough...

Turns out yeah no, you really do need 2, we think movement between gearbox and engine has caused it to wear, which damaged flywheel where the bushing sat.

The adapter plates also don't have a huge number of bolts holding everything together, especially at the bottom where their is the highest load.

Adapter Plate

Originally the gearbox bolted up to the adapter flange with 3 bolts (yeah I wasn't to impressed by that at the time, but a lot have been sold, so figured it would be ok), it's now been all modified to have 7, and more coming in from the engine side. We also managed to get one bolt at the bottom to go all the way through from the gearbox to the engine sump bracing (highest load). It won't be moving anywhere now...

We machined the flywheel to fix it up and put a proper bearing in it (2032 6 series - I think from b series), so that should also take some mis alignment better.

Engine mounting points, top of adapter plate sits above the top ones.

 

Gearbox Shifter Linkage
For any one interested, here's a good photo of shortening a gear lever linkage on an Honda S2000 AP1 gearbox to move it forward which i took while it was out.

 

 

 

 

[Hyperblade]

So I went for a shakedown with new clutch, which overall went pretty well.

 

I got a comment on one of my videos which I thought I would answer here as it's long and complicated and may interest others.

Quote

@cbdrift5864
Its the first time ive seen it driven in anger - this thing has some serious front end bite and sounds like its VERY enthusiastic Good job! From the looks of your dashboard your running a link ecu? Have you had this dyno tuned or just doing it yourself from the basemap? I have a jdm k20a on a haltech and im always curious to find/swap/get more info on tuning the engine on an aftermarket ecu - especially the cam control side of things which i spend a lot of time on.

Map

First up, I don't mind sharing my MAP so others can learn, It's for a LinkEcu G4X XtremeX and you can download it here 

https://drive.google.com/file/d/1-M1MXHnkJxH2A30xRdFbANd65I309fiA/view?usp=sharing

Tuning was done on a dyno by a specialist, so can't add any extra info there, other than maps were done for non vtec/vtec then it was worked out where the crossover made the most sense (4500 rpm).

Driven in Anger

The comment about it being driven in anger is a good observation. With the BEAMS in the car, I always felt the car was hard to turn in, and if I tried to be aggressive it would just understeer. That plus suspension valving issues, rear brake pads accidently being to aggressive, I never felt comfortable pushing it into a corner and so was reasonable gentle on the inputs to give lots of margin for error.

I have finally reached a point where we have made major suspension changes, lengthed wheel base, anti dive, shocks valved correctly, rear pads correct, and 30kg taken directly off the front end, where I'm finally feeling comfortable in the car.

However I'm still used to driving the car fairly gentle, which you would see in 2nd shakedown video from the same day on my channel. 

I got Brent who did the fabrication and who has driven a wide range of very fast race/rally cars into it for the 3rd shakedown video to see what he thought, he came away absolutely loving it the handling with no changes required to valving, or sways bars or anything major which we are both stoked about.

But he did have the following comments:

• Could do with Power steering (will explain why below)
  • Suggested Suzuki Ignis mid 2000's electric steering column as you can get the aftermarket modules to control them and they are small.
• Power is not what it should be in the top end (suspects down 20hp)
  • suggests shortening secondaries.
  • Suspects that they are to long and it's affecting scavenging at high rpm.
  • Whole logs/fueling needs to be looked at to see what's happening first before going down that route (but i need to learn how to read them and adjust them first)
  • Can easily see it on the dyno sheet, I've just lived with it for now as other priorities, but obviously something not quite right.
• The watts linkage should be lowered to improve the angles
  • this will increase grip on the rear end on exit of corners
  • Shakedown video 5 has it lowered by 15mm, which may have felt slightly better.
  • But have reach limit of adjustability, so need to lengthen the slots.

So the power steering is the interesting bit, the reasoning is that when turning for a corner the power steering moves quicker for less input, with the tyres I run if you turn in very quickly initially (you can back off afterwards) they hook and get a lot more grip then you would normally get.

I have never had the car in the state before where I felt like throwing it in would work or that I could safely control it.

So Session 4/5 I changed my driving style to try initially turning the wheel into a corner really fast.

Holy hell, it was an amazing feeling that i have never experienced before, it blew my mind, it was hard to compute how much front end grip it suddenly had and the rear just felt like it slipped slightly then hooked up too, it's hard to convey how big a difference it felt but felt like I had at least double the grip on the front end.

As you can see in the videos (4/5), I was missing apexes all over the place as trying to time it was really hard. And that's why power steering is recommended as it means you just turn into the corner and the natural movement will cause it to hook at the right time. Where as doing it manually is slightly slower and timing it is harder.

 

[Hyperblade]

So things have moved along rapidly in diagnosing the power loss. Lol

I had Dave at Dtech motorsport look over the tune and logs, as he has tuned more then a few honda K engines.

He was fully happy with the tune and thinks it looks all good.

Best guess at this stage is the intake manifold (Skunk 2 Ultra Street with 2x .5L plenum spacers) is to big for the stock engine.

He advised removing the spacers initially, but unfortunately they are also used to get the throttle body into a position where I can get the intake pipe around the oil cooler duct.

So onto moving forward with my ultimate long term (now short term) plan which is ITB's.

Aside from the power gains (hopefully) they will mean I can do a really nice cold air intake setup. The car was built with lots of space in that area to play around as I knew I would eventually end up at this point.

The recommendation from him is Jenvey Tapered ITB's as they perform the best.

https://store.jenvey.co.uk/honda-k20-ep3-sf51-taper-kit-ckha07-kit

I don't have the money for that, so over the coming months (in no rush for this) I'm going to backyard build a setup.

I have aquired Roman's BMW S1000RR 48mm ITB's and will 3D print manifold, trumpets, airbox, intake  to suit.

It's electronic and I currently run throttle cable, but I think I have enough outputs/inputs on ECU and can just buy an new linkage kit for my existing pedal (https://tiltonracing.com/product/600-series-throttle-pedal/), that it probably makes sense to keep it electronic and not need to put Idle up valve in.

Just going to be a lot more wiring i was trying to avoid. however the benefits probably outweigh the pain.

 

[Hyperblade]

So I acquired a 3d Printer a wee while back and I've been learning how to design and create prints, and I finally had a cool job race car job for it.

I needed to lengthen the Watt's linkage slots so I can lower the linkage to get more grip out of the corners, and it's really important they are straight to keep the axle alignment.

So I 3d printed an insert that I could then outline the slot and center point for drilling the end of it out.

Worked a real treat and allowed me to do a much nicer job.

 

So the big downside of using @Roman ITB's are they are electronic, this means 2 TPS sensors, 2 pedal sensors new relay to supply power to the B plug of the ECU controlled by the ECU, and repinning and moving connections around.

I had thought at the time when wiring the car about DBW and thought "na I'll never go DBW"

Sigh.

So knowing it was a pretty big job I thought it might be worth remarking the top of my current manifold as an interim measure, so with Daves help I designed up a new top cover.

But once I got to that point I could estimate the amount of material need to print it and I was also very concerned with the strength of it hanging off the side of the engine with throttle body with the only attachment being bolts threaded into it, I'm sure it's all solvable with time and effort, but I knew I wanted to do ITB's so scrapped that plan. It did give me lots of experience designing a complex shape which was good.

So onto the wiring job.

My goal was to rewire ready for the ITB's but still run the Intake manifold for now.

So I previously had these bulkhead connectors to the engine. And they are nice, however having had them on the car for a while I wasn't happy with them.

They are big, but more importantly they are impossible to work on e.g. good luck getting the very center wire out. They do make it much easier for when testing the wiring as they provide a nice test point for connections, but otherwise they are really ideal for if you had another engine with the loom ready to go and drop in, but in reality I'm not a big race team doing that I'm just a club racer.

So my goal was to create something that was more modular that I could upgrade/add bits over time easily.

So I designed this in Fusion 360, printed mockups got it to how i wanted it, then had it laser cut and then @ajg193 machine the slots for the clips on the back side.

It has standard Deutsch DTM/DTP connectors which were a bitch to work out how to retain, and they still aren't quite right (but that's to do with the clips on the back and I think I can do something better). 

But it will work for now.

Really happy with how it turned out.

It runs the following.

• 1: DTM12 (Front of engine) - Crank, Vtec, VTC, Oil Pressure, Oil Temp
• 2: DTM12 (Back of engine) - Exhaust Cam, Intake Cam, ECT, Starter, Alternator feed
• 3: DTM12 (Intake) - IAT, MAP, TPS Main, Fuel Pressure, ISCV.
• 4: DTM12 (Front of car) Transponder (can deal with lights etc in the future if required)
• 5 DTM8 (Gearbox) Speed sensor
• 6: DTM8 - Ignition Coils
• 7: DTM8 - Inboard Injectors
• 8: DTM8 - Outboard Injectors (not used at present)
• 9: DTP2 - ECU Ground to engine head
• 10: DTP2 - EWP
• 11: DTP2 - Fan
• 12: DTM4 - Can Lambda

Connector 3 has everything wired up for the ITB's as well so just need to change engine side to suit, also potentially supports variable velocity stacks (BMW S1000R comes with them as standard).

The idea is If I suddenly need a new sensor, I can easily splice it in to the engine side, and just run a single wire back to ECU.

I've also wired extra sensors into the steering columns for the pedal sensors and clutch in the future so they will be straight forward to add later.

Car is finally back together, It's running but I need to get it out on track again just to make sure everything is still good.

 

 

 

[Hyperblade]

ITB's Next Step

So the wiring is now done just needing testing.

The ITB's are a long term thing I need to slowly work through.

So while rewiring I pulled the manifold off so I could measure the ports and create a flange in CAD so I could slowly work on the manifold adapter.

 

So much crap on the internet

I did a lot of research on the internet around K20 ITB's and the port sizing, and can confidently say some manufacturers have both the port angle and port size wrong in their drawings (and that's their job to get right) Jenvey are definitely the ones who have it the dimensions right and that's backed up by feedback from someone who has tested a few and found them the best.

The jenvey ones are 48mm and everyone says that bigger then that is better (e.g. 52-54mm), but that ultimately depends entirely on where the throttle blade is positioned with the system.

e.g on the following diagram, A, B and C are all different plate sizes but really have no impact on the amount of air being ingested, however position does have an impact on response and driveability (closer to head being better).

So the Honda K20a CL7 port size is equivalent to 46mm throttle plate, which means if you put the throttle plate at B (48mm) you have accounted for the restriction it causes.

Jenvey looks to keep a straight section (no increase/decrease in size) of 46mm diameter for a length of 85mm from port to throttle body flange. I suspect that is because you don't want to take energy away from the air by trying to increase it's velocity just before it goes into the head, but could also be packaging etc. I'm just speculating here.

Additionally Jenvey have the port angle at 17.5 degrees, which when I measure is perfect for dead center (vertically) of the port as it narrows. Where as they others have 14/15 degrees which means the top side has a sharper angle then the bottom (which is flat).

So I can see why Jenvey perform, their experience shows in actually getting the basics right.

So they are the ones to copy as I can't afford their full kit

BMW S1000RR ITB's

Now the S1000RR ITB's are 53mm entry, 48mm plate and 42.84mm exit with a 3.5 degree taper on walls.

The exit is not good if port size is 46mm., however I'm confident I can bore them out with a little effort on the lathe to 46mm which would be perfect, I only need to bore 10mm into them, so don't have to worry about the throttle plate.

So I pulled them apart to see how hard it was and if it was possible to put each half on the lathe as is without having to completely dismantle the plates

Then they wouldn't go back together, a small bit of panicking later I was thinking I would have to dismantle them which would then mean they needed to be resynced.

But in the end I 3d printed a tool that now allows me to put them back together as they were from factory, so that's a massive win.

Bore spacing was the other worry the ITB's are roughly 83-82-83mm and the CL7 is 94mm, but it turns out that it's close enough that it will be fine.

 

So I can proceed with the ITB's and the design of the manifold to suit them.

My intention is to mount the standard Honda injectors into the manifold in the same place as the Skunk2 manifold (close to head) so that the transition of the tune to the ITB's is easier, then setup outboard injectors at a later stage (and just block up the BMW ports).

 

I'm not rushing this so it will be done when it's done.

Just in comparison here is what others have done to fit bike ITB's to a Honda K series.

https://www.jimbositb.com/product-page/throttle-body-adapter-plate-k24

 

https://danstengineering.co.uk/Honda-Civic-Type-R-K20A-Bike-Throttle-Bodies-Kit-GSXR1300-46mm-STARTER-PACK

 

Feel free to tell me I'm an idiot here; 

 

[Hyperblade]

The last club day of the season (I was out in first round so not worried about points) was coming up and it was looking wet 

So I thought it would be a good opportunity to get the the car out for another shakedown of the new wiring, and get some experience in the wet (I've never driven in the wet).

So first time out on a wet track in qualifying, i put it on pole...

Race 1

Turns out it goes really good in the wet. Allowing me to take my first win by 30s over 2nd place (he sand bagged a bit after he accidently hit pit limiter and saw me race off into the distance).

 

I managed a 1:42.5 which is I'm absolutely stoked with, that's only 7s off my dry time, and there was a little more in it (and that's with no changes from current dry setup). 

Also it confirmed that things like the wipers, lights and demisting (not running any, instead I'm using the heat from exhaust + radiator) all worked as expected.

The wet's i ran are 4 year old ex Michelin TRS, and are absolutely amazing, it really was like driving in the dry, they are super soft and it shows as another competitor who is normally slightly faster in dry then me had older harder different brand wets and was 12s off my time.

But also the weight of the car and it's handling characteristics show it's driveable/chuckable nature.

Race 2

Race 2 was a handicap race and I was 60s behind first pack, unfortunately the guy who came 2nd place in 1st race was 15s in front of me, and he picked his pace up to match mine, i did manage to get within 9s of him on a damp track (rain had stopped).

 

 

I came out of the day actually feeling excited to potentially race in the wet in the future, where as before I was very nervous, so that's also a massive win.

 

ITBS

Interestingly enough Tegiwa have just come out with some Jenvey ITB's to fit an EP3, they are very heavily compromised to fit in the engine bay, but do have some dyno charts.

Check out the dyno results and see what you think about the results and claims...

https://www.tegiwaimports.com/blog/?p=6297

Quote

A run of 251.1bhp with 161.1 lbf.ft meant the kit had improved performance with an additional 13.8bhp! Of course, dyno figures aren’t the be all and end all of everything, it’s largely about the experience on the road. As we all know, there’s not much that comes close to the noise of a set of ITB’s screaming all the way to 9,000rpm which is a huge factor when it comes to deciding what option to go down when tuning your K20.

 

Before:

After:

 

 

Let me know your thoughts in my discussion thread here:

 

 

 

 

 

 

 

 

 

[Hyperblade]

Small post to say a big thank you to @h4nd for coming round and getting an Oscilloscope trace of the Honda S2000 Speed Sensor (VSS) I've been having issues with (only taken me nearly 2 years to get round to it.) just waiting on Link to respond with their thoughts now, but have one potential solution we think should solve it.

 

Traces are below if anyone else needs similar in the future

 

 

[Hyperblade]

Haven't posted much as been working through annoying gearbox issues.

So in one of the races I was going into the hairpin and instead of selecting 3rd it went into 1st. 

In a honda that would have been engine gone, but as it's RWD it immediately locked up the rears and i had a bit of a moment, but no damage.

Additionally in the last race on the last lap as i crossed the finish line I attempted to go from 5th into reverse...

Not ideal...

S2000's run a reverse lockout where you have to push the gearlever down, for some reason it was already down when i went to go from 5th to 6th and so it smoothly tried to shift into reverse (to right of 6th).

The issues:

• The gearstick had always been slow to recenter if you let go, so while braking your trying to pick the center gate, rather then just letting it go then moving it straight forward.
• I probably forgot to lube the ball and the shift lever worked it's way down at some point over the laps so the lockout wasn't in effect.

Solutions

• Pull apart gearbox and double check its ok and put a put a stronger centering spring in and a install an upgraded detent kit (near disaster).
• Lube the gearlever, and get an 80% stronger reverse lockout spring made (works well)

 

Go back out, and just as i'm starting to do a fast lab in quali shift from 5th into 6th 4th... 

 

Well that's an even bigger issue, it literally just smoothly went into 4th no complaints and felt like it was going into 6th. 

Thankfully I had dropped the shift point from 8800 to 8000 to see if it would improve the overall laptime, so the engine only hit 9550rpm...

So some head scratching later, and pulled some of the upgraded detent springs out, which improved it from going into 6th 10% of the time to 95% of the time just sitting in the garage.

Went out again.

Unbelievably did it again, but it took until race 3 (good footage of me dodging a stalled car on grid as well).

So at this point this has to be sorted properly, the engine will not take much more of that.

Looked at hand position, concluded I hadn't really changed it between good and bad.

 

So the Issues

It turns out the detent kit was not installed as per the instructions (plus there was a last minute change to where a couple of springs were put then it wasn't tested), so that's a big issue on my side. However one of the springs puts an insane amount of pressure on the selector shaft (hence why it wasn't put in), to me it's just been poorly designed. So all upgraded detent springs removed, the center spring stays.
 

But one of the issues is that their is so small amount of movement needed to go from the 3-4 gate to the 5-6 gate.

What do i mean?
Here's the lever in 3rd,

Here's it in 5th

That and I think the kinematics of my position is not helping the engine/gearbox are on an angle in the car, engine more over to passenger (left) side, so as you pull gear lever back, you have to put more pressure to the right. I think the detent springs accentuated this with the amount of pressure they required to hold it across.
 

So solutions:

• Make the gear lever longer to give more throw between gates
• Remove the rubber isolator in it to remove any play in it.
• Find out the gear lever is also bent to the right, so straighten it... (this makes the kinematics worse)

 

So here's the inside of a Honda S2000 gearshift lever (aluminum sleeve has rubber bonded to it)

Spend time making up a longer shaft and end up with this

 

Throw wise it's going to slow me down a bit, which is a good thing.

It puts the lever height wise in the middle of the wheel, which is perfect.

 

So fingers crossed all those changes finally give me some confidence to be able push it hard again.

[Hyperblade]

So onto the next problem, I've had warped rotors for a while (low down on priority list), so i had them skimmed ended up with 0.12mm taken off of one side...

So went out again, car was braking amazingly, I could brake so later and with huge confidence, then they started warping again.

Ok need to sort that out as the rotors aren't cheap. They have also been hitting 600c+ which is not good for rotor or pad life.

I brought knockoff makita blower and I put it on the intake to the brake duct and measured where the air was coming out.

30kph straight in and 10kph straight out the rotor at the red circle, 0 out the vanes.

Granted in hindsight, its bloody obvious, but just hadn't really through through the ramifications and how bad it was.

Fundamentally the hat (GP4 Fabrications Kit) is just not designed correctly it should have no holes in it.

One 3d print later I had 

Testing that alone made a massive different to the airflow (naturally, it's so fucking obvious now) with air coming out 1/4 of the vanes.

Here's the old cooling, so while I'm there i may as well do the job properly and make sure the air doesn't come out the other side either as that could cause warping as well.

Multiple iterations later we have this

So after getting laser cut we end up with these

Clearance to disk will be adjusted if it requires it.

That's the simple part done, hard part is now ducting the air into it nicely/safely.

You can see the technical chat on that here:

Fundamentally current plan is to go with a PLA mould and fibreglass/carbon/Kevlar on top of it, then softening/melting it out, I have someone doing the work for me whos an expert in fibreglassing.



Heres attempt number one (carbon/kevlar).


He had a go at getting the Mould out but it put up a fight. So I've printed some thinner moulds to see if that makes it easier for him.

However I think if I put it in the oven at 70c it should melt out, so going to try that too, if it works it's very cheap and easy to make the moulds for complex shapes.

Now naturally while the front brakes were working amazingly (being straight and all) the rears are now starting to get up past 600c as there is not a lot of room to get air into them, 258mmx21mm disk with AP Calipers.

 

So work is proceeding to see what can be achieved with redoing the caliper mount and/or moving disk outboard for easier maintenance/replacement.

 

[Hyperblade]

So 5 months later I finally managed to get some time to myself to look at this.

It didn't help that i broke a 4mm tap in the last hole of the hats months ago, which certainly didn't put me in the right mindset to look at it (not to mention work/kids/life keeping me very busy).

I also decided to buy new rotors as I should have solved the cooling issues now and wasn't confident that another skim would be good for the old ones, the new ones were eye wateringly expensive.

So last weekend i finally got my head into the right space and got into the garage and had look at where i was and what i needed to do.

Decided i should stop pissing around with all the stupid ideas on trying to get the tap out and just get @ajg193 to mill out the tap which he promptly did trivially the same day. (yeah I'm an idiot that should have got him to do it sooner). Go check him out at https://www.goodengineering.co.nz/ he's awesome.

So this weekend got the new rotors on and the new plates mounted.

Came out amazing, clearance is perfect, so stoked with it all, the 3d printing really made this job doable for me, otherwise i don't think it would look anywhere near as good as it does (or be possible).

Next step is to 3d print some template to drill some locating holes for the carbon fiber ducts i have, then I plan to glue the ducts to the aluminum.

 

 

[Hyperblade]

Slow progress on this, but chipping away at it when i can.

 

Carbon fiber ducts are fitted up, I just need to glue them to the plate (for additional strength) which I will do as the last step.

 

 

 

[Hyperblade]

Next step is to get the hose to them, I want to redo this area so I have a nice transition from the spoiler inlet which is 2" to the hose which is 2.5" and I want to provide some better clearance to the wheel so it doesn't rub through anything.

So finally got out the 3d scanner and this is my second attempt at scanning the inner fender well (einstar via laptop with USB passthrough to desktop to test if i could do it remotely without dragging desktop with me...).

Pretty happy with it, and can really see the benefit of scanning an area when trying to fit something into it.

So prototypes are now being printed in PLA to see if the real world matches the virtual, weight of mount + duct is 130g (PLA) which i'm pretty happy with.

 

[Hyperblade]

Final versions of the front brake ducting have been printed in PA-CF (Nylon/Carbon Fiber) and fitted, came out great (It's super strong), The ducts have brass heat inserts for threading the bolts into which have turned out far stronger then I expected which is awesome. I have some new silicone hose coming as the aliexpress stuff i have is clearly below par and that's the front done all done.

Weight each side is 150g (including mount).

Them printing:

 

Onto the rear now, I've chopped up the already modified caliper bracket to get air into the center of the disk, will do the other side, then fit them and 3d scan and make something to guide air into them.