ECU Tuning by Volumetric Efficiency

[Roundy]

Hey Guys,

Recently it looks like the adaptronic has given support to tune by volumetric efficiency.

Are there any benefits to doing it this way?

Adaptronic are saying easier initial tuning, which i suppose makes sense as it isn't too hard to guesstimate a VE, and you enter the engine size and injector size.

I believe Autronic also tune by this method?

Cheers
Nathan

[oldcorollas]

Megasquirt has done that from the very start. it seems the intuitive way to do it, and i don't know why others did not do it before.

it does mean that you can just set say, WOT at 90-100% VE (100%VE is still an arbitrary number), and then work your way down with MAP/load.. and then tune from there.. also means you can use VE map estimators for different cam/engine combos.

it makes it really easy to swap injectors and injection settings, since the calculation from injector size to VE is done automatically (by calculating 100%VE pulse width), and then you can easily swap from say.. 1 injection event per cycle, to 2, or to alternating injection events etc, and still use almost exactly the same VE map.

it also makes it veyr easy to apply enrichments (or enleanments?) as a % rather than a ms pulse etc...

it makes it really easy to transfer between motors with different injector setups, and different turbo setups (since things tend to even out above 1atm).

[YLD-16L]

If I tune any more Adaptronics I'll tune by VE but I won't be bothering to change my one from injector mS to VE values, no benefit in doing so in my case as it's pretty much sorted now.

The benefits of VE tuning Vs injector mS tuning has been well described by OC already.

[oldcorollas]

just wanted to add, the VE estimator for MS is open to the public here

http://www.megamanual.com/v22manual/vetable.htm

plug in your details, and away you go.

anothe rmajor benefit of VE tuning is... your table and settings are largely transferable between different platforms , but i supose this is a negative for aftermarket ECU companies as it reduces the dependence on their brand or tuners?

The VE curve is assumed to follow the torque curve (which is not quite right, but close enough for a first approximation). This calculator is based on modeling your engine's wide-open-throttle VE with a quadratic equation of the form


VE = A*rpm2 + B*rpm + C

where:
A, B, & C are constants.
These constants are derived from the:
peak torque@rpm,
peak horsepower@rpm,
and the assumption that the peak torque occurs at the VE curve's maximum (i.e. the first derivative with respect to rpm is zero).
An assumption has been made about the VE required to produced 1 lbｷft/ci as well as 1 hp/ci. A further assumption has been made that compressor efficiency=60% for kPa > 100. Finally, a scaling factor has been used to adjust the WOT VE to lower kPa. The factor is derived from examples of working VE tables.

A limit of 5% per rpm bin drop in VE has been applied to prevent the extreme drop-offs that can occur at high rpm & kPa when VE is modeled in this way. As well, all VEs are restricted to be not less than 25% of the highest VE, to prevent unusable idle values.

See the source code for this document for more details on how the calculator works.

when i tuned my MS, i starte doff with high VE values everywhere and tuned down to good numbers.. with ms pulsewidths, it feels like shooting in the dark... with VE, you feel liek you get more of an idea of how the engine performs in difrferent areas.. of course you can get the same data by normalising your ms pulsewidths but...

the other thing is.. from a VE table, you can simply use the ms pulsewidth for WOT for your engine, and then calculate the ms required for the rest of the table (can do simply in Excel), so the VE table makes tuning more portable to every platform...

if you wan tto get into more nitty gritty, read thru (or skim thru) the tuning guide for MS, which details pretty much every part of it.
http://www.megamanual.com/v22manual/mtune.htm

the algorithms for the MS tuning are also available, so you can see how VE tuning is affected by different enrichments.

BUT, VE and % enrichment is still not perfect and is still an approximate method, since the actual requirements of an engine are wuite complex, even when thinking of a single variable, BUT, it is intuitive, easier to understand, and easier to port to elsewhere...


on a sidenote, changing the "req. fuel" parameter (theoretically required fuel for 100%VE per cylinder) will change your VE values... so maybe you might have 80% max VE, or 130% max VE in your table.. that doesn't really matter, but you can adjust the req fuel so that the %VE numbers make more sense (ie, up to 90 for most engines, or 100 or 100+ for developed motors)

"etc" says the beer

[Jason Broadhurst]

Not sure if it is noted here but i'll give a quick run down on VE

The VE fuel map basically calibrates the engines VE to a specific AFR/Lamda.

For instance if you were to set a target AFR of 15:1 for all loads and rpm (target AFR tab in wari/seukukuskuku), then tune the VE map so that every single load point ran 15:1 AFR, your VE map would be calibrated/tuned. This then allows you to change your target AFR table which in turn alters the AFR the engine will see, without adjusting anything else like the VE map.

It is very handy, you can change AFR's easily, without a retune. If you were to do the same thing with a ms fuel map, you would have to adjust each load point individually and check with a wideband.

Also remember that dead time/lag/latency is quite important. This gets a lot of people with PFC because they don't know what they are doing and adjust the latency to get the car to idle but it throws out the rest of the map.

Any physical change to the motor (cams, timing changes, intake, turbo) will affect VE, but it also affect a ms table so its no problem.

[MWP]

Not sure if it is noted here but i'll give a quick run down on VE

[edited]
Comments removed.

[wilbo666]

Hrm, maybe reconsider doing so until you learn the theory behind fuel tuning.
What youve said doesnt make much sense at all, and a lot of it is just wrong

It might not make sense if you're not taking it from an Adaptronic perspective... but Jason is refering to how the recently released Adatronic VE tuning sytem can work..

i.e

Now both Adaptronic ECUs have VE tuning. Download the latest e420c firmware and software (V2.000) from the downloads page. The graph below shows the results on one of our test cars (a Mazda MX5 NA BP). The process was that the engine was tuned by VE with a target AFR of 12.5, to achieve the bottom of the AFR curves (violet) in open loop mode at wide open throttle. The target AFR was then changed to 12.8 and then 13.0 for two more power runs, with no tuning in between. The flatness of the AFR curves, in open-loop, speaks for itself.

As a note I also appreciate this graph for its comparison of AFR & Power



Cheers
Wilbo

[oldcorollas]

MS does same thing but implements slightly differently.
the Req-Fuel is based on engine size, injector size, and arbitrary AFR.
if you change that AFR, it changes the Req-fuel, which will change the map as the Adaptronic does.

if you have wideband, then you can use the AFR target tables, which will calculate the amount of fuel based on the req-fuel target AFR and the table target AFR

[gixer]

i starte doff with high VE values everywhere and tuned down to good numbers.. with ms pulsewidths, it feels like shooting in the dark... with VE, you feel liek you get more of an idea of how the engine performs in difrferent areas..

Totally agree, I prefer tuning with VE, much better feel for what the engine is doing/requiring.

[MWP]

It might not make sense if you're not taking it from an Adaptronic perspective... but Jason is refering to how the recently released Adatronic VE tuning sytem can work.

Oooh, ok, thinking about it a little more, i see what he was getting at.
Sorry Jason

[gixer]

You can tune HPtuners software like that

[Jason Broadhurst]

VE is VE but yeah I was talking about wari/sekuku specifically

I missed your comments anyway MWP Funny enough I was out tuning

[RyleyMA61]

Thread necro. I have a 7MGTE and it's cool.

I am upgrading from an E11v2 to a PS2000. The latter supports VE.

I have a 97% complete base fuel map, in fuel injection time, ready to be imported into the PS2000.

- Should I bother with VE?
- If so, could I just copy a VE map from a 2JZGTE single turbo (provided with Haltech), given the 7MGTE is somewhat similar in volume etc?

If I go VE, is the correct process:

- Set up inj size, deadtime, target AFR map, etc, VE map
- Confirm by tweaking VE values that the target AFR values are being met using wideband

Is that it?

[Roundy]

The PS2000 may have an option to convert your existing map to VE anyway.

The reality is VE vs injection time tuning, you get the same result ultimately, but VE is easier on the initial setup.

But yes your idea for the basic VE setup is pretty much spot on, but probably need to tell it the capacity as well.

[trdee]

Adaptronic has a function which allows a ms map to be converted to VE and vice versa. Hopefully hsltech has it too?

VE tuning really is the way to go. It is way, way faster to converge to a decent tune than fucking around with injector ms. You will need to input engine capacity as Roundy said, along with injector size and number of cylinders. That is the bare minimum. I don't know what options the hsltech gives you but basically put in the maximum amount of info you can as it will help with the accuracy of the VE "estimate"