| Home | FAQ | News | Downloads | Pictures | Technical | E-Mail Archive | Members | Join | Search | Links |
| Home | FAQ | News | Downloads | Pictures | Technical | E-Mail Archive | Members | Join | Search | Links |
To see my original notes and updates from previous years, I refer you to the following pages:
Stuff for people to download:
List of recent updates:
January 12th, 2006 |January 14th, 2006 |January 18th, 2006 |February 13th, 2006 |March 10th, 2006 |November 22nd, 2006
It's all coming together now! I spent a good part of my Christmas break taking care of all the wiring the project needs. I ended up running something like 24 or 25 wires from pretty much one end of the bike to the other. The scary part is that it all works. Watch out for hot soldering irons - they tend to produce very nasty burns before you even notice. It's been about three weeks, and it's almost healed now.
Mick never got a chance to weld my tank together before Christmas, but he and Frank have been working on it the week. I saw it today on my way home from work, and it's going to look very sweet when it's all welded together. I'll have to clean all the metal filings out of the tank, and repaint it, but those are small prices to pay.
If everything goes according to plan, I'll have a go at getting the bike started sometime next week when I'm on leave from work. I expect that it will take some convincing to start, as the CBXs are always a bit cantankerous about starting after they've been sitting for some time. We'll see what happens.
Ah yes... The new subframe. I received an arc welder for Christmas, after procrastinating about buying one for months. After melting a few bits of scrap together, I went and bought some steel so that I could build a real subframe for the bike. As you can see from the photos, it's an interesting shape. Given that all it has to do is hold the seat on, brace the exhaust, and mount the indicators, I think it's a pretty functional design. I took the opportunity to raise the seat by about 20 or 25mm, as it really was a fraction too low.
So, it looks like this project is in it's final stage - I just have all the brain-bendingly interesting tuning work to do. Once I get the thing running.
| Image size | File size | Viking's comment | |
|---|---|---|---|
 | 800x600 | 77kB | The throttle bodies mounted on the engine. |
 | 800x600 | 45kB | Air temperature sensor and mounting. |
 | 800x600 | 25kB | Hot soldering irons will burn you. As I found out. |
 | 800x600 | 54kB | Ultra-high tech choke. |
 | 800x600 | 54kB | Spaghetti wiring. |
 | 800x600 | 63kB | Snazzy looking pod filters. |
 | 800x600 | 44kB | Viking hard at work soldering wires. |
 | 800x600 | 74kB | More spaghetti wiring, at the ECU end of the loom. |
 | 800x600 | 50kB | Subframe as it takes shape. |
 | 800x600 | 78kB | Completed subframe, after painting. |
 | 800x600 | 52kB | Throttle bodies and throttle cable clamp. |
 | 800x600 | 35kB | Throttle stop to prevent the throttle from opening too far. |
I picked up the tank from Mick on Friday evening. I must say that Frank is a damned good welder - the work was absolutely perfect. OK, the paint got ruined in a reasonably sized area, but I knew that was going to happen anyway.
I spent last night cleaning up the burnt paint, and discovered that someone had put a layer of body filler about 1mm thick all over the tank at some stage in the past. I guess this explains why the surface was so smooth. Anyway, I got the wire wheel out, and used the power drill to scrape back the burnt paint and body filler, all the way down to bare metal. Once it was cleaned up, I put a good coat of primer paint on.
Today, I cleaned up the primer and put a second coat on. I don't really want this tank to rust. It will be a cast-iron bitch to get another one...
I really, really need to thank my friend SteveL, too. He dropped around this afternoon, and was instrumental in helping me getting the bike running. His enthusiasm and problem solving skills are pretty well second to none.
Together, we put the tank back together (installing the filler cap is a prick of a job), then went about problem solving why the bike didn't want to run. First, there was a kink in the petrol line from the pump to the fuel rail. Then, the injectors weren't passing fuel.
We eventually found out that the injectors were clogged with gunk and not opening! Bah! Humbug! Luckily, we were able to carefully press the injector pintle back with a small screwdriver. We're talking less than a millimetre here, by the way! Working the pintle back and forth a few times seemed to work, as we started to get fuel into the inlet manifold. A few more moments of cranking, and the bike coughed, farted and roared into life!
I can't really describe how I felt... I am very happy that everything was working at long last, but depressed in knowing that I have a huge amount of work in front of me getting it tuned properly.
Overall, a great day today!!
| Image size | File size | Viking's comment | |
|---|---|---|---|
 | 800x600 | 46kB | The tank as I got it back from Frank, burnt paint and all. |
 | 800x600 | 54kB | A bit of cleaning and polishing later... |
 | 800x600 | 71kB | All back together again, and ready for tuning. |
 | 800x600 | 42kB | A strange view of the pump, showing it's base and mouniting bolts. |
 | 640x480 | 2800kB | An MPEG movie (MP4 format) showing the bike starting and running. |
Oh boy.... This EFI tuning is a lot harder than the manual made it out to be. Mind you, the manual is aimed at car drivers who are able to have a co-pilot handy to do the on-the-fly tuning. Us bike riders have a bit more trouble when it comes to tuning whilst moving.
To cut a long and painful story short, I think I have the base fuel map down pretty well. There are a few mild stumbles when keeping a constant throttle, but some datalogging and some thought should be able to sort something out. A session on a dyno is on the cards, too. Preferably one that has a wide-band Exhaust Gas Oxygen (EGO) sensor handy.
I just wish I had a working speedo - then I could see just how fast she does accelerate with the right fuel amounts and a healthy dose of acceleration enrichment enabled.
A word of advice here - if your fuel calculator says put in 'X' amount, then put in 'X' amount! Don't piss around trying to think you can out-smart it. You can't.
Ah yes, datalogging. I have a borrowed Palm Pilot that can act as a datalogger, I just can't get the data out of it! So it looks like I might have to strap my laptop to the bike somehow and go for a spin. I'm really not looking forward to doing that. Of course, having the laptop present means I can make quick adjustments on the side of the road. Hopefully.
If you do use hacked up throttle bodies, like I did, you'll need a little stopper to hold the butterfly valves open slightly at idle. Otherwise your engine will not run. You can quote me on that, if you like. *grin* See the picture below.
| Image size | File size | Viking's comment | |
|---|---|---|---|
 | 800x600 | 56kB | How to clean your Exhaust Gas Oxygen sensor - a careful burning with a blowtorch. Just be careful not to get it too hot. |
 | 800x600 | 52kB | Yet another bracket - this time to hold the throttle slightly open. |
 | 800x600 | 68kB | It's not elegant, but it works - although I got tired of plugging that cable in all the time. |
 | 800x600 | 52kB | The tuning software is Windows-based, so I have to run VirtualPC on my Apple laptop. |
 | 800x600 | 58kB | Datalogging, the hard way. Maybe I need a smaller laptop.... |
Hmmm.... Yes. Time for an update.
Speed-Density is dead! Long live Alpha-N! Yes, I have switched algorithms, and let me tell you, it's much better now.
I have the fuel map mostly sorted out (I hope! Now, where have I said that before?) for most of the lower half of the RPM range. Cruising at about 80km/hr in top gear is smooth as silk. It's still running a bit rich (going by the seat-of-the-pants) past about 1/2 throttle or so, which makes the engine sound rather lumpy and rough. Cold starting is interesting, it takes a couple of goes to get the engine to fire and continue running. Once it's running, though, and past that first 20 seconds of coughing and farting, it idles quite well.
One thing that I have noticed is that I have to change my riding style. No more can I come into a corner one gear too high and rely on the bike to slowly accelerate. I have to come into the corner in the right gear, at the right engine speed and accelerate smoothly away - if I don't, it just bogs down and goes nowhere as I open the throttle. Whether this is due to the incomplete tuning (including those pesky Acceleration Enrichments) I haven't been able to tell yet.
I'm beginning to have my doubts about the EGO sensor, too. It seems to want to indicate that the engine is running very rich, even when it's one touch away from stumbling because it's too lean. I guess that if worst comes to worst, I can always disable it and simply run in open-loop mode.
I haven't been doing much on the bike lately, I've had too much other stuff to worry about. I've leaned the fuel map out a bit here and there, and it seems to run a lot better. I still think it's running far too rich, but we'll have to wait and see.
The big news is that I have Raven booked in for some serious dyno time down at MegaCycle. With a wide-band O2 sensor, I should be able to get the fuel map pretty well sorted out without too many hassles. The advantage of working on the dyno is simple - I can hold the bike at a constant RPM and throttle position and get the map sorted out. No need to try and extrapolate backwards from datalogs, no need to try and remember what's happening as I'm riding, no need to watch out for cars.
The only worry I have at the moment is that she's burning oil (and a reasonable amount of it), and that's what's killing my O2 sensor. I don't really want to kill Ken's WB sensor.
I'll let you all know how things turn out...
Wow.... 8 months without an update.
OK, I got Raven on the dyno and a wide-band O2 sensor stuck into her exhaust pipe. (Ooops, sounds a bit rude doesnt' it?) Anyway, 2 hours later, I had a working fuel map.
It's a bit different to what I was expecting to see - the bottom half was running way too lean, and the upper half was way too rich. Although I think that the bottom end could do with leaning out a fraction now.
The downside to all this is that Raven has been parked under a cover since about Easter. Now that the EFI project is proven to work (and is working fine!) I no longer have the motivation to finish the job (and fix the blasted oil leak, along with getting the exhaust valve guides replaced). I've ridden her a few times over the months, and she is a fun thing to hoon around on, but the "fun" has gone out of owning her.
Perhaps it's time to find a new owner for her....