Thursday, 25 July 2013

Running lights

Running Lights

A high priority for me is to be seen particularly by SUV vehicles, they just don't look down low enough to see kit car drivers.

I found some Audi running lights they are 170mm x 16mm x 5mm thick. I bought these from ebay I could not find another product 5mm thick. The wiring was more or less in place as I provided for it in the loom.

They fit at the lowest point on the J15 and don't look out of place when off. When switched ON you would not ignore this car unless you are blind!

Off
Definitely ON!

References

Contents: http://sylvabuild.blogspot.co.uk/2012/01/sylva-j15-is-kit-car-designed-by-jeremy.html
ebay: Super Bright DRL Daytime Running LED's - MATRIX AUDI PORCH

Friday, 12 July 2013

Road trials

Road trials

Took it out on the road today, taxed, insured, registered. Fitted the aero screen  for better comfort needs more wind deflection. Not fitted side screens as I don't know how to get in to it with them fitted.

Good points

* The road holding and ride quality are very good better than expected.
* The gearbox ratios although not perfect are better than the Westfield getting 40mph in 1st.

Bad points

* Far too much vibration at certain speeds +/- 2500rpm, esp in town in traffic and can be irritating to drive. Vibration maximum at 50Mph in 5th, 30Mph in 4th.
* Engine hunts when moving in traffic. Can promote a kangaroo like response unless you are quick with the clutch. All the web searches indicate a problem with the ICV or PAS over pressure switch.
*  Keep forgetting indicators need a sounder to remind me they are on.
* When fitting the screen the rear view mirror needs relocated. Further brackets etc to be made.
* Needs running lights
* Sharp projections below the steering lock


Mods so far

*  Added cover to bottom of steering lock to remove sharp edges (that should have been a problem at IVA but was overlooked thank goodness).
* I have subsequently checked the PAS switch and indeed it was in the wrong state -  Obviously I have no PAS no kangaroo throttle when wire connected to correct ground.
* Modified the offside and nearside engine mounts and introduced a rubber sandwich. This has had a dramatic effect on the resonance at 2500rpm. It is not perfect but much improved.
* Modified location for rear view mirror
* Added a Cobra sun visor to act as a wind deflector.
* Mapin pizzo buzzer to indicators to stop me signalling permanently.


RVM changes

Engine mount sandwitch

Wind deflector


Proposed nearside sandwich mounts
 
Nearside sandwich

Revised engine steady
 

Contents: http://sylvabuild.blogspot.co.uk/2012/01/sylva-j15-is-kit-car-designed-by-jeremy.html

Wednesday, 3 July 2013

Chasing Lambda failure

Chasing Lambda failure

The IVA failure of excessive lambda ratio of 1.06 vs 1.03 limit needed fixing. In addition to that the car is not performing that well being very hesitant between gears and the revs holding up too long. The IVA man wants less than 1.03. It took several trips to the local MoT station to try and get a hint of what was going on. At first I thought the exhaust manifold was pulling air due to a faulty gasket. The gasket was faulty but the high lambda remained stubbornly high despite my efforts changing the gasket.

The MoT tester John at Regent Motors suggested testing the inlet manifold, using a can of liquid butane and squirting it around the inlet manifold region. This is apparently used by some to find air leaks in the system. I took a can of camping gas, the gas that is used in one of those cheap stoves. I fitted a spray head from a WD 40 can. The spigots are a different size but a 4mm drill soon fixed that. The can of butane spray was then applied to the inlet side of the manifold, plenum and throttle body.

An area with a suspected crack was identified as a possible failure in the inlet manifold casting. I gave it a big squirt, this gave rise to a temporarily rich mixture that resulted in the engine running slower then speeding up again. I knew right away a new problem area had been located.

Getting the manifold out from the area behind the seat meant separating the manifold and plenum.  Five of the plenum button screws sheared the hex head detent. After drilling out the screws the broken flange was taken to Aliblast Services in Linlithgow for repair. He did an excellent job making a good penetration weld on each side. The next steps were to reassemble the manifold and plenum and return to the MoT station.

Crack viewed from below

Crack once dismantled

5 out of 10 screws drilled out on plenum

Welded flange 1

Welded flange 2

Return to the MoT station

Further tests revealed the emissions for Lambda had improved but only just enough but sitting on 1.03 limit. More work needed to be done to give a margin at the IVA station.

I decided to have a look around for further faults on the basis I don't think there is anything wrong with the various O2 sensors however I have fitted another new one in desperation. However, I found if I stuffed a wet towel up the exhaust I get fumes coming out of both the joints in the exhaust system. This can be a path for introducing Lambda type emission problems. I am not totally convinced as sealing the manifold made no difference. However plodding on, I made up copper gaskets for the two joints. I made them out of 1mm thick copper strip designed for moss prevention on roofs. I then packed the intermediate space with jointing compound. I then clamped it up tight with the existing clamps in both cases. A side effect is the exhaust no longer pops when the engine is revved up. The wet towel no longer  shows up any leaks.

Copper gasket

Diagnostic tools

I have been using some affordable diagnostic tools to help with this stage of the process:

1) An application on my Android phone called Torque and the associated Bluetooth code reader -  ELM 327. It can be used for plotting your progress around the track.

2) Sealey Lambda tester/stimulator VS925

Both these tools can reveal how the management system is dealing with the issue of it's closed loop performance. The first tool has an application called Fuel air status. It monitors if your system is achieving a closed loop control over emissions.

The second tool monitors the voltage on the Lambda sensor and tells you if the oxygen sensor is operating over the correct range.

I also tried a trick I found on YouTube for testing lambda sensors but in combination with the Sealey tool. If you connect up the Lambda sensors on the bench to the Sealey VS925 and heat the sensors with a  blow lamp (blue flame) you can monitor the display just like I did in the car. Initially the one that was unresponsive became responsive after a few minutes glowing red hot. In other words you can restore lambda sensors with a blow lamp if they become contaminated.


Hot news 3/7/13  @11:22

References:
1.01 Wow even at tick over!
 

In summary

* Leak in inlet and exhaust systems (4 places)
* New Lambda sensor
* Fuel tank vent valve
 

IVA retest

Passed! now for Build Up Inspection.