turbo build 2023

[Joseph]

All that looks ace 

[gsxwill]

Keep the build going, sounds interesting 

[Breadman]

All of the space I thought I had is rapidly disappearing.

☹️

 

[Breadman]

New Bosch stick coils fitted into original Suzuki spark plug rubbers and they seem to be held quite tightly despite the weight and extra height of the coils.

Edited July 28, 2023 by Breadman

[Breadman]

Finished all the pipework and fittings for the fuel system.

Put some fuel in and gradually pressurised the system. Got up to 2 bar and no leaks.

I was very happy that the modified fuel rail and injectors had no leaks.

I fully tested the system up to 4 bar and still no leaks.

Very, very  happy.

Edited August 2, 2023 by Breadman

[Breadman]

Fitted the oil cooler and oil lines today.  The steering lockstop will have to be altered now as the forks contact the oil cooler bracket......but thats a small problem to be dealt with at a later date.

Also, I cut the tube for the plenum chamber and this needs to be finished so I can then start fitting the various sensors and at least make a start on connecting all the ecu wiring together.

The electrics for fuel pump, ignition, oil pump, etc will be on a seperate loom.

This bike is going to be ugly and very rough around the edges. The goal is to get it running.

 

[Gixer1460]

14 hours ago, Breadman said:

This bike is going to be ugly and very rough around the edges. 

You ain't lying there LOL! Do you have shares in Goodridge hose and fittings? 

Personally, I don't see the point in a lash up just to get a noise out the pipe, only to rip it all to pieces to re-do it and make it pretty! Why not build it how its going to be, build it right the first time and only build it once? Yeah it takes a bit more time and impatience gets in the way but doing it twice isn't quick either?

[imago]

20 hours ago, Breadman said:

I'd stick some support onto that 'T' piece. With the weight of it + the AN fittings, and the hose lengths that'll whip about like an expensive escort at a politician's birthday party. If/when it either cracks, loosens or pulls the connections there'll be oil everywhere.

Edited August 3, 2023 by imago
typo

[Duckndive]

4 hours ago, imago said:

I'd stick some support onto that 'T' piece. With the weight of it + the AN fittings, and the hose lengths that'll whip about like an expensive escort at a politician's birthday party. If/when it either cracks, loosens or pulls the connections there'll be oil everywhere.

Personally i would have the TEE at one end or the other "less fittings needed" 

[clivegto]

One I did earlier, worked alright.

[Breadman]

Bit more progress.

Power connected to the distribution board. All relays functioning as they should. Oil and fuel pump, A/F gauge operating. Engine turns over and kill switch functions. Oil pressure warning light connected.

So just the ECU sensor wiring to finish off and I can connect the laptop and set up the parameters to get it running.

I will keep the ECU wiring seperate from the switches and relay stuff.

Wrapping the looms will happen later.

[Arttu]

I would recommend using the ECU for fuel pump control.

[Breadman]

Yes, I have the white wire from the ECU connected to the fuel pump relay.

[Breadman]

As suggested by a few people, I will reduce the quantity of oil lines. Rewelded the top engine breather outlet.

Wiring is almost complete. Just the a/f meter to wire in.

Arttu has helped me with a wiring solution for this.

I hooked the laptop up to check that all the ecu, fuel pump and sensors were all functioning.

Next week, I will do a final check of all connections and nuts and bolts. 

Recheck all the laptop parameters are correct again and then I can check the oil pressure before giving it some sparks and fuel.

[Breadman]

All wiring completed. Everything works as it should.

Time to hook the laptop to the ecu.

Got fuel. Got sparks. But the motor refused to start. It did try though.

On the plus side, I have plenty of oil pumping around the motor. This was a concern after my crankcase modification to allow the crank bolts to clear. I obviously can't check oil pressure properly until its running.

So I have several batteries charged ready for next Monday when it is time to start altering some settings in the ecu and hopefully get it running.

This is new territory for me. I've read the manual (several times). Had some good advice given but its me that has to change the settings. Its going to be interesting.

I need to access the plugs to check them so I will have to leave the petrol tank off and extend the 2 fuel lines. Despite having quick releases, it's still time consuming and a bit of a pain.

[Breadman]

It,s alive and I am over the moon.

Tested the injectors again . All injectors are putting out the same amount of fuel.

Megasquirt has a test mode so you can fire each one individually or all at the same time.

Next to test was the coils. These all gave nice fat sparks .

So I went to fire it up and all it did was cough and splutter.

Removed the spark plugs and the plugs were wet so I had fuel. Checked each coil again as I cranked the motor and 1 and 2 cylinders were good but no spark on 3 and 4. Time to dig into Megasquirts many ecu menus and settings. Finally found something that wasn't right, corrected it and it and it fired up. Very happy and relieved.

The bike idles nicely about 1000rpm but definitely needs a better fuel map at part throttle and above.

As its a freshly built engine, I'm being careful.

The facet oil pump needed a bit of persuasion to prime but is now taking the oil from the turbo back to the engine.

Now I can take off the wiring loom and get that tidied up.

Big thanks to Arttu for his help. Top man.

Edited September 26, 2023 by Breadman
spelling

[Breadman]

[Breadman]

All the wiring has now been wrapped and just needs a bit of heat proofing in certain places.

Swapped the Lithium battery for a gel version. This seems to handle the constant charging and engine cranking much better and I can use the lithium on something else.

I fitted a AN-10 pipe to the turbo and this has a 20cm distance to the facet scavenge pump. I have discarded the oil tank that I was given and as long as I remember to run the pump briefly before starting and after stopping the engine, it seems to be working fine and no oil smoke on start up.

I have been trying to get the injector deadtime as accurate as possible for the Subaru injectors. Looking online has given me some varying figures but none of the figures quoted were taken at the 13.2volts that Megasquirt 2 uses as a baseline. 

Im all new to this ecu programming and whilst a ballpark figure will get your engine running, it just makes sense to find out the exact deadtime for my ecu and injectors (time it takes for the injector to open and close when its told to).

YouTube to the rescue and a solution.

I started doing what I was told in the video.....and.... the fuel rail o-ring on the injector split. Swapped the injector for another one and then the alternator stopped working so gotta fix that before I can start the test again.

The A/F meter has also lost its 12v power from the ecu and I cant understand why. It works and functions fine when connected to an alternative power source but it would be nice to solve the puzzle.

The end of November seems about right for some testing and I have been trying to find a dyno shop which has availability and knows their way around the Megasquirt/Tunerstudio software.

I have decided to treat the bike to some new Suzuki clutch plates but this comes at a horrendous cost !!

I would hate to have it slip when I've got it on a dyno and waste the time and money spent getting there.

Lets hope nothing else breaks before then.

 

 

 

[Arttu]

12 hours ago, Breadman said:

I have been trying to get the injector deadtime as accurate as possible for the Subaru injectors. Looking online has given me some varying figures but none of the figures quoted were taken at the 13.2volts that Megasquirt 2 uses as a baseline. 

Im all new to this ecu programming and whilst a ballpark figure will get your engine running, it just makes sense to find out the exact deadtime for my ecu and injectors (time it takes for the injector to open and close when its told to).

YouTube to the rescue and a solution.

I started doing what I was told in the video.....and.... the fuel rail o-ring on the injector split. Swapped the injector for another one and then the alternator stopped working so gotta fix that before I can start the test again.

The A/F meter has also lost its 12v power from the ecu and I cant understand why. It works and functions fine when connected to an alternative power source but it would be nice to solve the puzzle.

I assume you are firing the injectors with the test mode at different pulse widths and measuring the resulting flow? That's the correct way to determine the dead time. Just be careful that the battery voltage stays stable during testing. Your injectors are large enough that the deadtime settings make some noticeable difference. But still small enough that it isn't absolutely critical to get it exactly correct. 2200cc/min injectors on a 1000cc start to be slightly tricky

There is a fuse inside the ECU for the lambda sensor feed. If you have accidentally short circuited the supply wire the fuse might be blown.

[Breadman]

The injector test involves measuring the fuel at various pulsewidths and then plotting onto a graph which I found on the Megasquirt forum. It does seem a lot of work to get the deadtime but I am sure it will be worthwhile.

I will check the ecu. Thanks for the info Arttu.

[Breadman]

Arttu's prediction about the fuse was correct.  Thankyou.

The AFR meter has a separate inline fuse of the same rating but that didn't blow ! 

I have now thrown away my cheap Amazon fuses and bought proper ones.

Fitted the new Suzuki friction and steel plates back into the clutch basket and finished it off with the FBM lock up clutch (diaphragm version) which I had acquired with the smallest weight of nut and hex bolt on all 4 of the clutch arms.

New wheel bearings were fitted to the Bandit rear wheel and I now need to measure and make some spacers up for the Bandit swinging arm so it fits in the Slingshot frame.

Now all of this has been done, next step is to start the bike again and use the autotune for the warm up enrichment....because I really am in new territory here and need all the help I can get. Have fitted a longer exhaust pipe with the Lambda sensor in a better position to aid the autotune facility.  Even though I intend to use just a short exhaust as you can see in my earlier pictures, I am liking this extended version. Needs a flapper cap on the end to stop the rain if I keep it.

My ageing laptop has also decided to stop talking to the ECU, yet another problem I have had to correct.

 

 

 

[Breadman]

I managed to get another laptop after the last one stopped functioning. Plugged it in and all was good again.

Unfortunately, 2 days later, I plugged it back into the bike and relays starting clicking and then it all went dead.   

Got out the multimeter and after a lot of checking through the loom only to find that the VREF output pin from the ecu was only showing 2 volts and this was the same voltage I was getting out to the TPS, MAP, MAF sensors. It should have been 5 volts.

So I had got a short circuit in the wiring and this had damaged the ecu.  BUGGER.  Only one thing to do now and that was to get it back to Arttu where he weaved his magic and fixed the problem, wrestled with the Finland customs and finally got my ecu back to me. Thanks. 
 

 

[Breadman]

Not wanting to repeat this again, I have ripped out all of the wiring and built a whole new, and simplified, loom and have also used so much cable protection and insulation to be 100% sure. Thankfully it has all been tested again and everything is working as it should. I haven't tried to run the engine as yet so that is work in progress at the moment. I have also got rid of the Innovate A/F gauge and I will now run a Spartan controller with the Bosch O2 sensor and just rely on Tuner Studio to display A/f settings and data logging once I can start riding and tuning the bike.

I have finished fitting the gsxr rear wheel, and made spacers to allow the bandit swinging arm to fit the slingshot frame. Bought an offset front sprocket and this all now seems to be in alignment. Currently making some different foot peg hangers as the original Suzuki ones didn't suit my riding position with the wide,straight handlebars as there was way too much weight on my arms.

[peter1127]

yeah making a loom is a lot of work. Pays off to start from scratch!

[Breadman]

With the wiring now reworked and insulated beyond what is probably needed, it was time to get the bike to start and run after my various problems.

Yes, it started but something else wasn't right. Massive overfuelling problems.  So I called upon Arttu to inspect the datalogs.

There is so much info from the ECU and I don't know what I'm looking for, it's just so confusing.  I am becoming more familiar with what is going on and how to interpret it especially when you have some help.

Back to the fuel problems. I checked through the wiring as Arttu had asked me to do and the map sensor seemed to be the culprit. The sensor was reading 1.6 bar all the time and was just launching fuel into the engine.

I drained the oil out of the bike and there was certainly some fuel in it. As the bike hadn't really run or been under load, I'm sure no damage has been done.

Took a recommendation by Clivegto and got some Westway 10/40 oil. Definitely cheaper than what I was using.

A quick fire up of the engine with no map sensor problems and the datalog showed that the fuelling was still a bit rich.

With the Megasquirt, you have to let the warm up enrichment phase finish before it transitions to your main fuel table at a given engine temperature. I shut down the fuelling some more and I now have a steady idle. I am very happy with this. Just to have the bike running for 10-15 minutes after spending over a year of my life getting to this point is just  so good for the soul.

Brake caliper refurbishment and installation is next on my to do list as well as some further attempts with the mapping.