Gotta love these sorts of conversations. Great Thread.

20% ethanol in my car again. After 100km without it, I noticed a decent increase in torque, probably from the leaning effect of the ethanol (same effect as the blitz ECU tune leaning things out)

Lol..
One little wrinkle I've come to realise with this blend, not that it's overly important, but worth mentioning, is the closed loop mode EFI adjustment.
Normally the O2 sensor will correct the EFI mixture based on exhaust emission O2 content, and because the ECU is adjusting Lambda value, the ratio of the mix isn't that important, as you'll still get close to optimum AFR's.
Now the slight wrinkle. Because the Ethanol component is oxgenated and exhaust flow is increased (proportionately) a stock narrow band O2 sensor is possibly going to see a slight increase in O2 levels, and read it as a lean mix, adjusting short term AFR's slightly rich. Over time the ECU is going to learn that condition and adjust long term O2 adjustment accordingly.
If you reset the ECU, these values will be lost, and it will start over, but nothing will  change.
 
So what does that mean.. The closed loop (idle/low load/cruise) is going to be slightly richer than optimum (economy) EFI settings. I think this is where some of the increased responsiveness is coming from, and that feeling of revability.
Down side is your going to use slightly more fuel in closed loop.
Upside is in open loop your leaning over rich AFR's and getting more power while using less overall fuel, well in comparison to output because you are injecting the same volume of fuel. Also it could be considered something of a fail-safe in relation to going lean, for some unforseen reason. And of course your getting more responsive performance.
How it all balances out I don't quite know.
I suspect it depends on how you drive, and how often your into boost. But as I said, I still think it's only marginal, and possibly the higher output at any point would result in a lower throttle opening in closed loop. You'd have to drive exactly the same way on 98 and 80/20 to get some real world idea..and that's hard.
 

/\ not hard set boost to desired psi and make sure the boost gauge stays on it lolll

antilag?
Lol. either way. i'm hooked... and 98 just smells noticeably horrible by comparison now :P 

mmmm

MuMan
Now the slight wrinkle. Because the Ethanol component is oxgenated and exhaust flow is increased (proportionately) a stock narrow band O2 sensor is possibly going to see a slight increase in O2 levels, and read it as a lean mix, adjusting short term AFR's slightly rich.

You seem to be implying that the oxygen content in the ethanol is not involved in the combustion, and becomes an exhaust by-product.  This is certainly not the case.  A stoichiometric combustion of ethanol and oxygen results in the formation of the same compounds as stoich hydrocarbon combustion, ie. CO2 and H2O.  The proportion of these compounds will vary between fuels but the common factor is that if the combustion is stoich there is no excess oxygen.  It doesn't matter how the oxygen is supplied to the reaction, if it's the correct quantity it will all react to form the above mentioned compounds.  The oxygen content of ethanol is the primary reason that a lambda of 1 represents a lower AFR than petrol.
 
The O2 sensor will only see a rise in exhaust gas oxygen when the mixture is leaning out, and the response of the ECU will be exactly the same in terms of long and short term trims as with hydrocarbon fuels (ie. petrol).  In the case of blending E85 at low ratios with fuel, as per the topic of this discussion, the AFR will initially be lean.  The ECU will learn this and correct during closed loop operation, eventually transferring this correction to the long term trims.  End result is that if the ECU was supplying lambda 0.7 at full boost on petrol, it will learn to do the same on E85 blend, unless this results in the injector duty cycle maxing out.

I think you might have misunderstood what I was getting at cogs. I covered most of what is going on during boost & heavy load conditions, where existing mapped values are dictating EFI. The additional O2 content is one of the main benefits and source of additional power..introducing un-metered O2 into the combustion process, much the same as N2O does without the need to add additional fuel.
 
I'm only refering to closed loop, where the narrow band O2 sensor is likely to see additonal O2 content in 'unburned emissions', not O2 from an imbalanced combustion process. The narrow band O2 sensor as you know, not being very sensitive outside actual stoichiometric, will signal this to the ECU as a lean condition. Regardless of how long this AFR adjustment takes to end up as a long term trim, it's going to be ever present..without an ECU2 or some other means of closed loop adjustment.
As I said, I don't think it's any major issue, and could be something of a benefit in some ways. I'm seeing it in the AFR trace on 80/20 compared to the initial 98 trace after a reset, so it's coming from somewhere, and I think that's the source.
Have you tried this stuff yet? Seeing your running a gen2 ECU with the provision to log AFR traces, it would be interesting to see the result. I'm keen to see what is happening to AFR's prior to, and after opening the T-vis on a gen2 ECU. I'll explain why later if your interested.
 
BTW this 80/20 is a safe mix in relation to AFR's up to 14+psi at least, and I suspect 16 would be no different, I just didn't want to go past stock boost levels without some means of monitoring on the donor car. On the gen3 it is well down into the 10's even up to 18psi..without any trim.

MuMan
I'm only refering to closed loop, where the narrow band O2 sensor is likely to see additonal O2 content in 'unburned emissions', not O2 from an imbalanced combustion process.

This is what I thought you meant, but I'm unclear as to whether you were meaning pure O2, or other oxygen containing compounds such as CO, NO2, etc.  If there's pure O2 in the exhaust gas then either the engine is actually running lean, or the combustion conditions are poor and the reaction is incomplete.  Even running an 'odd' fuel, a well developed engine like the 3S is unlikely to reach closed loop conditions where any significant amount of incomplete combustion is occurring.  Any oxygen rich compounds in the exhaust will not be detected by the O2 sensor, so the low level emission byproducts that contain oxygen molecules will not affect the closed loop correction.  I'm not completely certain, but I don't think O2 sensors will even detect O3 (ozone).

Have you tried this stuff yet? Seeing your running a gen2 ECU with the provision to log AFR traces, it would be interesting to see the result. I'm keen to see what is happening to AFR's prior to, and after opening the T-vis on a gen2 ECU. I'll explain why later if your interested.

I've been keenly following this thread and keeping an eye out for places selling E85, to no avail :(  And yes, I'm very interested in any thoughts you have about the effect of T-VIS.  Unfortunately, the T-VIS on my MR2 is intermittently sticking closed so I'm not really able to give any reliable input at the moment.  The car desperately needs some TLC and I'm just not getting the time to get to it.

I see where your coming from. Forget about incomplete combustion and normal O2 trim.
Probably this is about a very minimal issue, but here's what I'm basing my thoughts on.
 
The mix as such isn't an issue as far as lambda 1 is concerned, but because there is an increase in VE, there is an increase in exhaust volume. This means there is some corresponding increase in Co2 and CO, also HC's. Now part of the emissions generally lumped under the heading of HC's is unburned fuel displaced between the piston crown and cylinder wall to the top ring. This volume (minimal as it might be) is ejected as unburned HC's along with any other by-product.. difference in this case is it's a source, but not necessarily the only source, of additional O2. 
The O2 sensor will detect some rise in O2 emissions and send a voltage to the ECU effectively reporting a lean condition. Of course the ECU will react by adding fuel until the O2 sensor again cycles through stoich and set a trim..which eventually is averaged and set in long term trim etc. etc.
Bottom line is it's there to a detectable degree that isn't consistant with , non Ethanol fuel.
No real problem in itself, in fact as i said can be an advantage, but spending long periods in closed loop is going to cost some (?) level of reduced fuel economy.
unless other gains balance it out.
  
 

OK, I see what you're saying, but for a couple of reasons it won't have the effect on O2 sensor readings that you're predicting.
 
Firstly, the trace HCs that remain unburnt in regular petrol are the ones with the largest molecules.  They take enormous amounts of heat energy to vapourise, have a high flash point and as you said will become trapped in the cool regions around the piston crown and never reach their flash point.  This effect is most prevalent in a cold running engine and is the reason warm-up enrichment is needed, ie. more fuel is needed so that the combustion mixture containing only the smaller, more readily vapourising components of the petrol is still stoich, while the larger molecules remain unburnt.
 
In the case of ethanol, it is a small and very reactive molecule with a relatively low flash point that vapourises readily.  Even in a cold running engine it will reach its flash point and combust right to the coolest parts of the combustion chamber, leaving no unburnt emission.
 
Secondly, as with the combustion by-products containing oxygen, the O2 sensor will not detect molecules containing oxygen in their molecular structure.  Even if pure vapourised ethanol was sprayed at an oxygen sensor it would not detect the oxygen.  Oxygen sensors only detect O2 molecules.
 
Edit:  You also referred to an increase in volumetric efficiency, which I'd dispute!  VE is a purely mechanical factor, using a different fuel doesn't magically allow the engine to pump more efficiently.  The fuel's higher energy density may allow more power to ultimately be produced, but this comes with precisely the same amount of displaced air.  Exhaust volume will be significantly higher due to the greater thermal expansion rate of the combusting fuel, but exhaust gas mass per combustion event will only increase marginally (by exactly the same amount as the mass increase in supplied liquid fuel, or around 4% in the case of petrol to pure E85).