Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Kevin Wood]

Has to be auto i'm afraid Al 

That does make any engine transplants (and tuning work beyond chipping and changing cams) quite difficult, unfortunately.

Whereas manual gearboxes can be pushed quite far beyond their original spec, automatic gearboxes are very much designed for the application. Their clutches need to cope with the torque output of the engine in question but still engage progressively to give a smooth shift, so they aren't normally engineered to cope with much more torque output than the original engine develops, as there is a balance to be struck. Witness the well documented problems with tuning automatic TD Omegas where the AR25 box is retained instead of replaced with an AR35.

Add to that the complication that most modern autoboxes are electro-hydraulic meaning they need an ECU to control them, and that ECU needs to be tightly integrated with the engine ECU.

You'll be fine going to 3.0 cams and chipping it, but any further than that, you really need to transplant a more powerful engine, automatic gearbox and control electronics in one go, IMHO. The MV8 project, in fact, if it ever materialises.

[aaronjb]

Unless you're happy with a properly old fashioned all mechanical auto box, of course, like a two speed Powerglide

On a serious note, there are aftermarket transmission controllers available for the 4L80/85: http://www.transmissioncenter.org/compushift_4l60e_4l80e.htm

They're good for 440lb/ft & 460lb/ft respectively.

[cem_devecioglu]

programming them requires dyno sessions which are also costly

Not necessarily.  I tuned my Megasquirt on the road with a wideband lambda sensor.

The advantage of this approach is that you learn an awful lot about what makes an engine work.

first, you have Kevin Wood handy


second, most tuners prefer initial session on the dyno for the car to run because tuners prefer biggest injectors possible which requires creating a map from scratch    after the car reaches "normal" working state , fine tuning done on the road

and a problem with that approach is that those cars will never be economical for daily use as tuners prefer low afr ratios to prevent detonation and high flow injectors squirt more even if you cut the duration times

[Kevin Wood]

first, you have Kevin Wood handy

Yes, but, despite that, it worked.

second, most tuners prefer initial session on the dyno for the car to run because tuners prefer biggest injectors possible which requires creating a map from scratch    after the car reach "normal" working state , fine tuning done on the road

True. Most tuners would prefer you to pay by the hour to use their rolling road.

I started out with a rough ignition map which was what I was using with the DCOEs that were previously on the engine. I had no fuel map whatsoever, in fact, I wouldn't be surprised if I was the first to run a megasquirt on a Zetec engine. I also used Jenvey individual throttle bodies running Volvo 940 Turbo injectors, so it was a bit of an odd combination anyway.

My approach was to calculate the fuelling requirement for the engine power I expected, turn that into an approximate injector duration and calibrate the whole system using that as a baseline for 100% VE. I'd observed that around 30% VE at idle is typical, so I shot at that and had it idling OK with a bit of tweeking. I then increased the revs off load and got it able to rev unloaded to the red line cleanly. I then "interpolated" the valves in the map from unloaded up to 100% VE and tweaked the acceleration enrichment until it'd rev cleanly. That's about all I could do without hitting the road.

Fortunately, I had a short downhill stretch of road before I could turn onto a main road and drive for 10 miles or more. I gave the closed-loop fuelling a lot of authority to correct the fuelling (by up to 50% IIRC) and set off. It drove like a dog, of course, but if you held constant speed and load for long enough for the closed loop fuelling to adjust you could watch the correction and decide how far you needed to adjust the map to correct it. A few manual corrections of the map and it was perfectly driveable. After about 30 miles it was almost perfect - better than the carbs I was running before, at any rate. It had a little more tweaking in the following months, but more than 10 years later it's still running perfectly!

Maybe I could get another few percent out of it on a rolling road, but I know that the tune probably varies enough with temperature and altitude / atmospheric pressure at the moment that that would probably be futile. Maybe one day I'll tune the correction for these well enough that I'll have a go at fine-tuning the map...

What I would say is that tuning the fuelling on the road is easy. You can tell through the "seat of the pants dyno" when you've hit the sweet spot and readings from a wideband lambda sensor can give you confirmation of that. Tuning ignition timing is much less easy, and I reckon a dyno would help greatly in getting that completely optimised.

[Kevin Wood]

Unless you're happy with a properly old fashioned all mechanical auto box, of course, like a two speed Powerglide

On a serious note, there are aftermarket transmission controllers available for the 4L80/85: http://www.transmissioncenter.org/compushift_4l60e_4l80e.htm

They're good for 440lb/ft & 460lb/ft respectively.

True. Actually, the Megasquirt guys were working on a transmission controller about the time I decided I needed to get a life outside that project.

It would actually be very nice to be able to tweak the Omega's autobox. I reckon you could make it a much more involving drive.

[05omegav6]

Has to be auto i'm afraid Al 

Then you won't have a car to drive unless you go for a much stronger box... Which makes it a wasted exercise.

[Webby the Bear]

Seems like there's sooooo much to take in to account. its actually a mine field.... and thats even if the bottom end of the v6's could hack it.

thing is it looks pretty darn straight forward on things like ''mighty car mods''. the hard bit being the wiring of the ecu and fabricating manifolds.

they shoved a turbo on an mx5. there was no mention of ''we've calculated we can do this'' so wonder if that'll start knocking quite quickly 

[aaronjb]

The MX5 is known to have a pretty bulletproof bottom end and takes very well indeed to turbo or supercharging - there's plenty of tried & tested aftermarket kits for that already

[05omegav6]

Thing is Webby, a 1.8 MX5 has a power to weight ratio of about 140bhp per tonne, and a manual box. To achieve this in an Omega you need at least 270 bhp. And if your saying that it must be auto, then you're looking at 300+ bhp to overcome transmission losses. And if you're going to look for that sort of power, then you're needing a much stronger box, and ideally a manual...

[Webby the Bear]

yeah fair enough guys.

i just think i havent thought through the implications... of which there seems A LOT!!!!

[aaronjb]

There's an old saying..

Speed costs money. How fast do you want to go?

[Webby the Bear]

 

That seems very true!!!!!!

i suppose it doesnt have to be a miggy (especially with all the talk of it not coping with FI).

I had an idea of getting something like a spares or repairs VW Passat TDI and fixing it followed by doing some turbo mods i.e. upgrading turbo just so i can get used to the process and pitfalls.

all things that go ''choo choo'' lol

[aaronjb]

If you're going to learn how turbos work for the sake of learning, I'd learn on a petrol engine..

The requirements on a diesel engine are very different due to the lack of throttle plate (well, ok.. some have them but let's not complicate things just yet), vastly different compression ratio etc.  While it is true it's very easy to get more power out of a TDi generally, you're much better served learning on a regular petrol engine.

My suggestion would be pick up an MX5 for ~£1000-1500, do the welding it'll need (practice with your new toy!) and then look for a cheap secondhand turbo kit for it, put that on, see how it all works and plumbs together when you're not trying to learn everything from scratch and you'll also end up with a very, very quick road legal gokart to have fun in 

[chrisgixer]

Has to be auto i'm afraid Al 

Quite right Webby. Quite right.

Of course, there is one engine with an auto strong enough to take a lot of extra power. Isn't there, gentlemen. ....ahem. Cough cough.

Not cheap though, looking at 4k with box and related engine management gubbins. Will probably have to be imported from the States unless your lucky.

[aaronjb]

Yes and if Webby goes that route he'll also be faced with all the steering mods and chassis mods that you've been deliberating over for months

Let's start simple for Webby's first learning project, eh?