Prawn and BigAls A3 Track Car

Sadly, my window motor video used to be the dullest, but then it got interesting because the window broke again right after the video was made!
 
T
daz-20vt said:
also I know you have a lot on but are you any nearer to poly windows. or plastic panels if anyone makes them.
I want to get serious with my car weight. gonna start grinding **** off soon. anything not needed is out. hoping to save another 50 kg somehow
Click to expand...
there are peolple In uk that make
 
You'd think so wouldn't you?

Plastics 4 performance want LOTS of money for a set, and the other companies who can do them don't seem all that keen on replying to emails despite a few initial leads.

I think the Greek set that Nick and Stacey were looking at were considerably cheaper
 
  • Like
Reactions: Miguel_S#
Why not buy the raw material and cut them out?
 
  • Like
Reactions: daz-20vt
I've been thinking diy
so are you doing 3 or 5 when it comes to it nick. you have said both! lol
I know it's not a priority just interested
 
hopefully none lol.
Sorry I tease!!
they say 40% less but that's no help without the original weight. also they do 2 3 and 4 mm thick
 
I've just sat and stared at that image thinking 'what the ****** hell has he changed on the car'....


....Then the lights came on
 
  • Like
Reactions: bigal 1, s3 Hoggy, Lavis89 and 1 other person
:blackrs4::racer:

Morning folks!

I've been fairly quiet in here lately.

Plenty going on, just not had time to post about it :)
 
So, where do I start....

Last week I took a week off work, because quite simply, I needed it!

Saturday was spent doing runs to the tip and general house stuff, Sunday was spent in Kent seeing Bob and Sarah, and new arrival, potential 2032 F1 world champ Peter v- cool little dude!

Monday I carried on the relaxing theme and went to Silverstone as a passenger, to spend the day with Rich, Nige, and the rest of the northloop crew – thoroughly enjoyable!

Tuesday was spent with Victoria, exploring the wonders of Hampshires finest countryside, and then the delights of CostCo :laugh:

Then onto Wednesday, and time to do some work!

First up I removed the air dam and bumper, and rolled the car outside to drain the coolant. At this point, I thought it looked particularly purposeful, so it seemed like an ideal time to take some pics!





And before I start any work, a current picture of the tired and tatty engine bay that’s seen over 3 years hard abuse with very little cosmetic attention:



Old rad out, and the new rad in, fan pack swapped over:



I also wired in a new EGT sensor, as mine had died, yet again! No pics of this stage, as it’s not interesting!

All back together, and test driven for what would hopefully be the final time on ME3.8!

 
  • Like
Reactions: Sootpig
By being very naughty :laugh:

Playing to see if the 'full throttle shift' function worked. Not advised! Mainly because I'll get told off :racer:
 
Then it was time to start on the ECU. A job I’ve been putting off for a while, as something always seemed to pop up, but I’d booked this time off specifically with this in mind, and I was determined to make some progress on it at last!

First up was comms, if I was to get anywhere with this, I’d need to talk to the ECU. This was pretty essential to check over the configuration and run through the recommended pre start checks before blindly turning the key and starting it up.

A quick trip to Novatech sorted that with this USB-Serial adapter:



I whipped the scuttle and wipers off, unpugged the stock ECU and removed it, along with the cage it sits in, then very loosely offered everything up, and plugged in the ECU using the plug n play adapter:



Before powering anything up, I unplugged the coils, ICM, and all 4 injectors, because if anything is wrong within the ECU it’s possible to fry any of the above components, or worse, the ECU itself.

I ran through all the pre-start checked of the map setup and ECU config, as per the emerald PDF guides. All seemed well.

It was then time to plug everything back in, and turn the key for the first time to see what would happen.



It runs!!!!

Ultra slow idle granted, and making a few odd noises, but this was genuinely the very first turn of the key and first fire up, and I’m very happy with it!

I had nightmares about installing everything and turning the key and nothing happening, and not knowing where to go with it, but with it running straight away, it’s easy to tweak and move on from here.
 
  • Like
Reactions: A3-Lee, StaceyS3, Rainbird and 5 others
Great work there Nick, That's a good start. Now it's just a case of moving forward as you say.
 
  • Like
Reactions: Prawn
Knowing that the engine was running, I set about mounting the ECU in a slightly more permanent way.



Space was limited under the scuttle tray, and there wasn’t access to get a drill in there to drill any holes to create new mounting points, so I decided to re-use an existing stud that had previously held the stock ECU cage in place.
I made up a little alloy bracket from some C-channel, and covered it in some 8mm thick neoprene to isolate the ECU slightly from vibrations



I then bolted the ECU to this, and offered it in place to check it fitted where I wanted it:





It fitted well, but there was a large void under the far end, and it was open to twisting under it’s own weight, which in time would almost certainly crack the bracket.

A slightly odd, but seemingly suitable solution, was to stick a large pad of neoprene foam under that end to support the ECU:



With that sorted, I set about wiring in the wideband 02 sensor output from my AEM guage. The AEM outputs a 0-5v signal, which is calibrated against a table supplied by AEM to correspond to a range of AFR values. From this the ECU knows exactly what the exhaust AFR is at any time, and of course, uses this for fuelling. This wide band capability was one of the primary reasons for going over to the standalone ECU in the first place :)

The green wire seen here has come from the wideband controller, and gives the ECU the 0-5v signal. This is wired to Pin 35 of the Emerald plug, and switched on in the user defined inputs of the ECU configuration.



At this point I decided to try starting it again, and making sure it was running correctly on the wideband:



More success!

Incase the video isn’t too clear, here’s the live adjustments screen shown whilst the engine is running:



Ignition can be trimmed using the + and – keys, with fuelling trimmed using keys 2 and 3, pressing enter to save the trimmed value to the map. Then when you go to the fuelling table, the trimmed values you’ve altered are shown in red so it’s really clear what you’ve altered.

Live and Target AFR are shown in a bar form as well as numbers, making it pretty easy to see how well that’s keeping on track, and then speed, load, and all associated temps and pressures are shown down the right hand side.

Lastly before closing it all up, Bill suggested via FB that it would be wise to shield the ECU from any water which might get in past the scuttle. My mounting system had already lifted the ECU up off the deck, so no pooling water would effect it, but it wasn’t protected from any water that may drop down from the scuttle, so I created this:



It looks pretty basic, but it extends beyond the ECU by about 50mm at each end, and will deflect any drips well clear of the vitals should any water find it’s way in behind the scuttle:



It was pretty secure the way I had slotted it into place, but I added a zip tie around the whole lot to ensure it stays put long term:

 
  • Like
Reactions: Alex C
Last up before the scuttle went on, I secured the map sensor, and routed all the wires neatly and zip tied them to the loom so they wouldn't catch on anything under there.especially with the wiper motor and arms spinning around in that area. I also routed the ECU comms cable through the bulkhead into the cabin to allow mapping on the move from inside the car without trailling cables in through the window

Seeing as page 188 seems to be the page of uninteresting videos, I thought I'd thrown in another scorcher for your viewing pleasure.

This is probably the most interesting of the lot!

 
Any Idea how hot it gets there, Is it not possible to mount the ECU inside the car? .It seems to be quite well protected from water/ dust ingress being IP65 and the connectors IP68 .But I have had a look on the emerald site and can't find anything that states the maximum ambient temperature.
 
Where's the picture of wiring the launch control switch to the steering wheel and video of said launch control..... :whistle2:

Did you have to reconfigure the start up tune as presumably they didn't have a base map for a ko4 TFSI turbo on a 1.8t with 440cc injectors?

When's she going to Bills? Any other changes to make before this?
 
I am still waiting to speak to my boss, who is never around, so I can confirm a day off and get back to Bill about booking in a date!

I'm going to spend the next week or two working on maps on the road, because I really want to fully learn and understand it all fully, then I'll be heading to Bill to put the car on the dyno and be judged for my first attempt at mapping, then hopefully, after some inevitable **** taking, he'll take pitty on me and work some Badger magic on it and make it fly :D

Then @Gops can send me some of those nice stickers he had made ;)
 
  • Like
Reactions: Gops and A3-Lee
Yay for progress. Looks a really cool setup, Nick. The geeky side of me is pretty wet for this right now...

antwan64og said:
Where's the picture of wiring the launch control switch to the steering wheel and video of said launch control..... :whistle2:

Did you have to reconfigure the start up tune as presumably they didn't have a base map for a ko4 TFSI turbo on a 1.8t with 440cc injectors?

When's she going to Bills? Any other changes to make before this?
Click to expand...

Reply based on little to no knowledge and experience:

At startup you're off boost so the turbo should make negligible difference to any pre-existing map and although the injectors aren't stock, I'd have thought it'd run albeit a little slower/lumpier than normal (as seen in the video). From there it should be a case of tuning and adjusting the map to suit, though at startup/idle that's fairly straight forward, unlike through the rev range/boost/power band etc.

Hopefully that's not too far off the mark, though I'm sure I'll be corrected soon enough if it is!
 
antwan64og said:
Where's the picture of wiring the launch control switch to the steering wheel and video of said launch control..... :whistle2:

Did you have to reconfigure the start up tune as presumably they didn't have a base map for a ko4 TFSI turbo on a 1.8t with 440cc injectors?
Click to expand...

The base map I was sent was for an AGU ko4 with 386cc injectors. I had originally thought it would be more suited to the 630's, so I fired it up for the first time on 440's at 4 bar, and whilst it did idle, it didn't rev all that cleanly or nicely.

I swapped to a 3 bar FPR and it cleaned up a fair bit.

I then swapped to the 630's @ 3 bar, and changed the scaling in the map to suit and played about with trims a little until I got it to idle and rev fairly cleanly.
 
  • Like
Reactions: antwan64og
oh how very original an update!

Man I don't understand most of this emerald stuff but I can't wait to see it finished. (though the beauty is constant adjusting I guess)
really pleased you man bet your like a kid in a sweet shop!
 
  • Like
Reactions: Prawn
antwan64og said:
Where's the picture of wiring the launch control switch to the steering wheel and video of said launch control..... :whistle2:

Did you have to reconfigure the start up tune as presumably they didn't have a base map for a ko4 TFSI turbo on a 1.8t with 440cc injectors?
Click to expand...

Rainbird said:
At startup you're off boost so the turbo should make negligible difference to any pre-existing map and although the injectors aren't stock, I'd have thought it'd run albeit a little slower/lumpier than normal (as seen in the video). From there it should be a case of tuning and adjusting the map to suit, though at startup/idle that's fairly straight forward, unlike through the rev range/boost/power band etc.
Click to expand...

Luke is pretty spot on. It'll idle (just about) at anything from about 10:1 through to 17:1 AFR, so a pretty broad range will get it running, them from there you can work out which way it's wrong and adjust accordingly.


After an initial test drive, it was clear that the base map wasn't ideally suited to the engine, but was enough to move the car around. I was casually chatting to Eddie (Toyotec) on Friday night, and he kindly offered to take a look over the base map supplied by Emerald and make 1 or two changes to try and make it more suitable for my 630's.

The AFR target map at this point looked something like this:



I had originally assumed, that I'd write the fuelling table as per above, and those would be the target AFR's at each given throttle position.

Of course, this doesn't account for MAP pressure, and I overlooked this.

In actual fact, the way the ECU works is a fuelling map based on TPS % vs RPM, and then to this table it applies a compensation based on the MAP (Manifold Absolute Pressure)

The MAP compensation table looks like this:



So for example, at 1 bar of boost, the engine will need 110% more fuel than it would with no boost, it would also need 10 degrees less timing, and an AFR richer by 0.6.

Whilst it's entirely possible to map the emerald as MAP vs RPM, doing it as TPS vs RPM and using a MAP compensation, means that adding or changing boost in future requires little additional mapping work, assuming the compensations are set up correctly. it's a pretty cool system!

What all this does mean though, is that if for example, I was at 1.1 bar, the map I originally thought was giving me 12:1 AFR at 6k, will infact be targetting 11.4, and if I up boost to 1.3 bar, that AFR target would reduce further to 11:1 AFR with an AFR compensation of -0.1 requested at 1.3 bar.

Confusing stuff! right?

The ignition table is equally complex to get your head round at first.
My current ignition map requests 28 degrees of advance at 100% TPS at 6k rpm, which sounds high, but this is before a MAP compensation of -10 @1.1 bar, so effectively, that becomes just 18 degrees of advance.

Once I got my head around this, I realised that the original base map targets of 13:1 AFR and 28 degrees of timing didn't seem quite so worrying after all!

I set the fuelling feedback table to 'adaptive', which means that as you drive it looks at the fuelling, and the ECU decides on changes required to try and get close to the target AFR.

After 15-20 miles of driving, I had zero corrections suggested, and the AFR's wern't THAT close to target at all. Clearly something needed tweaking.
 
A bit more research, and I found that the window for adaptation to occur had been set to be too narrow, and unless the car was right in the narrow window at each load / rpm site, no adaptation could take place.

Looking at a few sources, i decided on new values for this, and opened the window right up



The speed and load integers in the above table were changed to 64, which is how far away from the centre of a load cell the car can be whilst adaptations take place, and the 'AFR target within' value was increased to 0.6, giving the ECU a better chance of latching on to a load cell and selecting an appropriate correction.

I went for a test drive, very gently around town, and immediately the car felt better. I read the map from the ECU, and straight away the AFR corrections table was starting to be populated.

I headed off to a quiet, long, straight stretch of road, with a friend driving the laptop, and went through a little routine I devised to try and help the car adapt quicker.

Using 3rd gear, starting at 2000rpm, I held the revs steady using my left foot on the brake, and cycled through low, mid, and full throttle applications, staying at each throttle site for around 2 seconds.

I did this all the way through the revs right up to 6k, then pulled over and checked the AFR corrections table.

I think this pic is from the second round of adaptations I did:



Several big numbers around 15. After the first adapations I had numbers in the low 20's, then these came down to ~15 on the second round, before coming down again on the 3rd cycle:



Once you have some corrections recorded, you can select to apply individual changes, or highlight the whole table and apply all changes:

 
i wont have to do anything at this rate.. :p keep it up prawny :)
 
  • Like
Reactions: scotty_24, desertstorm, Golfyste and 1 other person
Street tuning like in the old days.

On my civic I used to use a Gtech, I'd make changes and do a pull then look at the "gtech dyno" screen and check the plugs, lol.

At least you have a wideband and EGT input. Keep up the good work!

What are you doing for idle control?
Emerald can't operate the stock weird stepper motor in the throttle body or can it?
 
Exciting stuff mate. I've got to say, i much prefer the sound of megasquirt mapping process. MS3 does it via RPM vs MAP on a 16x16 table, and you basically set the table up to cover the range of boost you'd ever see from the turbo + a bit. Then you can set the N75 boost control to whatever. Once you have the ignition table somewhere near sensible, tuner studio will adjust the fuel table according to the AFR feedback.
 
Awesome work dude :).
Very tidy install indeed!
And awesome to see it up and running straight away!
Looking forward to seeing how it develops!
 
badger5 said:
i wont have to do anything at this rate.. :p keep it up prawny :)
Click to expand...

That's the plan Mr. B! It'd be awesome if I could get it somewhere half close on my own, to minimise work for you, and also i really enjoy learning :)

antwan64og said:
At least you have a wideband and EGT input. Keep up the good work!

What are you doing for idle control?
Emerald can't operate the stock weird stepper motor in the throttle body or can it?
Click to expand...

I have fitted an idle screw to the AGU throttle body.

What you do is set it up when hot to idle with very low ignition advance, and set the hot idle on the screw, this then gives the ecu some air to play with when Coke and the chance to raise cold start idle using ignition advance.

It's idling perfectly at 850rom currently, but I need to play around with the idle advance again as above to get it to lift cold start idle a bit better as I've currently got quite a bit of ignition on normal idle and not enough from the screw.


Sootpig said:
Exciting stuff mate. I've got to say, i much prefer the sound of megasquirt mapping process. MS3 does it via RPM vs MAP on a 16x16 table, and you basically set the table up to cover the range of boost you'd ever see from the turbo + a bit. Then you can set the N75 boost control to whatever. Once you have the ignition table somewhere near sensible, tuner studio will adjust the fuel table according to the AFR feedback.
Click to expand...

It's Interesting how the same problem
Can be approached differently isn't it?

The emerald can actually be setup to do MAP vs RPM (speed density) exactly as you do in mega squirt, like this:



I can't say why, because I don't fully understand all the pros and cons of either method at present, but I do know that John at Emerald, and I believe also Bill, both choose to use the TPS vs rpm setup with MAP compensation when mapping Emerald.

What I can say is that the MAP compensation setup seems to work really well, and allows changes to boost with very little in the way of additional mapping.

After my hours test driving and mapping on Monday night I was able to use a full range of throttle openings, right up to WOT, from 2000-7000 rpm at 1 bar of boost. AFR tracks the target map very closely, and amazingly, even on part and light throttle application it drives really well with no jerkiness or hesitation that I've found so far.

To test how well the MAP compensation worked, I wrote a very quick open loop boost control table, interpolating from 0% dc at 1000rpm to 50% dc at 6000rpm.

This was purely to test that the n75 output as working, and I wrote it to build progressively to ensure I didn't end up with a crazy spike doing damage.
I was very pleased to see, that after it spoiled up, boost rose gradually to 19psi at around 5k rpm, before talking off slightly at the top end.

EGT stayed good, and AFRs tracked the target map really well even at the higher boost, so the compensation is clearly working well :racer:

I think the timing is still VERY conservative, as EGTs are slightly higher than I'm used to, although nowhere near dangerous at all. Egts naturally drop with increased advance, so that's something to look forward to when I start to look at the spark table :)

I am booked into Bills 2 weeks today, so I've got 2 weeks to get it as close as possible before I am to be judged for my efforts :D

However, it rained today, so I pulled it all apart instead :laugh:



Having a garage is AWESOME!

:racer:
 
  • Like
Reactions: Alex C, jojo, Rainbird and 2 others
The MS3 does speed density, MAF, alpha N (TPS vs RPM), etc but afaik it doesn't have the option for a MAP correction table, as generally if you have a boosted engine you use speed density, though I just about got my MS3 idling, then removed it again ready to add the expansion board to give me knock detection, and fully sequential injection (so i can do trim/cylinder)

No where near getting mine fully installed, let alone tuned, so like you, i have much to learn about turning, and still to figure out stuff like whether I'm going to run closed loop AFR correction, should i fit an EGT probe, setting up closed loop boost control, etc etc... All the fun you can have when going standalone ECU.

The more pops, bangs and flat foot shifts the better :D
 
audi upgrades
Thats interesting that Bill and John both like to use hybrid alpha N as load on turbo engines.

Some people mainly advocate hybrid alpha N for N/A motors with wild cams and supercharged engines if needed as boost is pretty linear with throttle.

i learn something new everyday, the emerald tuning suite looks quite nice.

Sometimes tuning works well just by how it "feels", I tweaked the carbs on my bike till it felt better and got I it dyno'd it was spot on.
 
You must log in or register to reply here.

Similar threads

J
A3/S3 Track car...
Replies
26
Views
8K
Prawn
Prawn
jsmotorsport
audi a3 8L track car
Replies
6
Views
4K
Prawn
Prawn
Marzo_NO
My A3 quattro track car
Replies
10
Views
4K
badger5
badger5
Sootpig
Sootpig's A3 track/drag car
Replies
166
Views
20K
Erikn89nl
Erikn89nl
M
S3 or A3 Track car?
Replies
40
Views
4K
Murf
M
Top
Back