Help with building your Speeduino, installing it, getting it to run etc.
User avatar
By Kinetic Research
#45595
Thanks for the suggestion, I had not considered the GM V6, they do have really good outputs. I stuck with SOHC to assist packaging as the front head is very close to the passengers shoulders but the GM heads look quite compact.

The McLaren F1 was originally going to have a Honda V12 but Honda were not keen to develop it, so I get a bit of a kick out of using a Honda powerplant. The J35 engine is 122kg with oil and coolant, vs the LFX's 157kg. Downside of twin engines is the weight penalty, so 2 x 35 kg is a big difference. The J35 with a cam grind and head porting has easily made similar power.

My budget is quite low for this project, so as I already have both engines I won't be changing spec at this stage. However the Tbox running twin engines was an idea I was originally going to install in a Mazda 787b replica. I would still like to build a racecar, but a roadcar makes more practical sense and is hopefully less lethal.
Last edited by Kinetic Research on Sun Aug 29, 2021 7:41 am, edited 1 time in total.
User avatar
By Kinetic Research
#51710
Unbelievable how time flies - thought I would have the engine running by now. Design is done and being submitted to LVVTA, not a lot of progress on the build to date, however that is all due to change over the next few months. Got the 3D prints of the exhaust manifolds cast in HT stainless steel. Just need to be machined and bolted on, sort the test engine frame and speedy setup then get some long overdue flames belching out.
Attachments
exhaust manifolds.jpg
exhaust manifolds.jpg (136.88 KiB) Viewed 4253 times
exhaust stub.jpg
exhaust stub.jpg (101.14 KiB) Viewed 4253 times
User avatar
By Kinetic Research
#52777
Got my build approval from LVVTA at the beginning of August so am now able to make some physical progress on the build. Shown is my measurement method for the 60-2 tooth gap and how it relates to the TDC on No1 cylinder. Printed out a 60-2 tooth chart and glued it to the civic flywheel using a precision whittled piece of PVC electrical conduit the same diameter as the pilot bearing and hammered it into place with a precision 4" x 2". While the glue was curing I rotated the engine to the 2 tooth gap position and then measured the crank sensor signal until it went low (from 3.3V when no tooth is present to 0v when it detects the leading edge of the first tooth in the correct direction of rotation) and then aligned the printed pattern with the precision bent piece of welding filler wire - so that now each printed line is inline with the falling edges of the CKP signal. Tomorrow will superimpose the cam pulse train over these CKP positions (glue should be set by then) and determine all the logic of the pulse train so that synch can be coded appropriately for cylinder 1 or Cylinder 5 after the tooth gap (since firing order is 142536).

From this photo the trigger angle for a J35Z2 appears to be 324 degrees (since each block is 6 degrees wide - tooth #1 is 6 blocks or 36 degrees before TDC)
Attachments
Mechanical method for measuring pulse train
Mechanical method for measuring pulse train
ckp pulse train.jpg (368.99 KiB) Viewed 4004 times
User avatar
By Kinetic Research
#52799
Dang it, thwarted by a dud CMP sensor. Since the timing belt and pulley have to be removed to replace it I will set up first runs with wasted spark & batch injection. Currently standing at the base of a very steep learning curve, so might as well have as many things simple to start with as possible.
User avatar
By Kinetic Research
#54005
Not much test engine progress over latest lockdown, however build is underway. Decided to make the rear end panel plug and fibreglass tooling to prove out a few ideas and get an idea for costs and work involved. CNC cut polystyrene foam blocks were epoxied to an MDF backboard, coated with epoxy and microballoon fairing compound (eases sanding the final layer). Currently detailing the fine polyester filler before the epoxy /glassfibre cloth reinforcing layer is applied, then more fairing and sanding before hibuild primer goes on and tooling is made from it.

Lots more engine related stuff coming soon, first I will have to enclose my DPscope/picaxe PCB scope and get to understanding the wiring setup and testing everything out.
last coat of epoxy.jpg
last coat of epoxy.jpg (91.62 KiB) Viewed 3381 times
Last edited by Kinetic Research on Thu Dec 16, 2021 2:41 am, edited 1 time in total.
User avatar
By Kinetic Research
#54278
I totally underestimated how long it takes to sand something like this to an acceptable level, but that is what learning is all about. Back to the J35 setup after the tooling is made from this plug.
Attachments
soft focus.jpg
soft focus.jpg (155.93 KiB) Viewed 3153 times
User avatar
By Kinetic Research
#56294
Can I please have some help clarifying my understanding of this circuit from my honda service manual please. As I interpret it the stock ECU supplies a 5V reference voltage over a pullup resistor on the CKP and CMP signal inputs which drop to 0V when the hall effect sensor grounds the signal wire.

When measuring the signal wires of the CKP and CMP hall sensors independently on the bench I am getting confusing results. I apply battery positive to (12V - yel/blk wire on terminal 3) and battery negative (0V Brn/yel terminal 2) however on the CKP blue signal wire I get 3.3V when no tooth present and 0.1V when "tooth" present. The CMP does not give any voltage on the signal wire with either tooth present or absent when similarly powered. Given the way the sensors are implemented in the circuit I was expecting the signal voltages to behave the same, and concluded that my CMP was duff since it is outputting 0V, however the new one imported from Honda Japan is exactly the same. I do not have another CKP sensor to compare, but would appreciate any advice on how to test and integrate these stock sensors to my 0.4 board (using the stock pins 24 for cam signal and 25 for crank)
Attachments
honda j35z2 ckp cmp signal.jpg
honda j35z2 ckp cmp signal.jpg (32.97 KiB) Viewed 2594 times
Last edited by Kinetic Research on Sat Apr 02, 2022 10:02 pm, edited 1 time in total.
User avatar
By dzener
#56296
Hello, very interesting project. Following from the beginning.
Your sensor needs a 5v PULL UP on the signal wire, as it works with grounding, you need something that keeps the voltage at a high level when you don't have a tooth reading on the sensor, as soon as it identifies a tooth, the signal goes down.
The speeduino already has a jumper for PULL UP, for testing put a 1k resistor between 5v and the signal of the sensor.

I suggest testing if this sensor works at 5v, if there is a problem and sending 12v to the arduino will damage it. I use a similar Denso sensor and it works fine at 5v.
User avatar
By Kinetic Research
#56299
Thanks for the advice, will setup a test circuit on a breadboard as described and check the voltages produced. Good idea to try driving the sensor circuit at 5V, safer option.

Hi, I am trying to assign Signed values to the x-a[…]

Vr Conditioner Noise when cranking

New version 202305 don't fix the issue. Now after[…]

blitzbox

I've finally figured out why MAP and Lambda weren'[…]

Hello, I bring news!! Let me tell you that after […]

Still can't find what you're looking for?