180 Degree L28ET Intake

[Gollum]

Actually, the cooling effect would be better seen POST blower. Think about it, why do we use inter coolers (or after coolers if you need to call them that) and not pre coolers? The higher the temperature differential the easier it is to absorb the energy. What this leads to is the reality that a bulk of the fuel in these carbed supercharged engines doesn't get atomized and provide it's cooling effect until down stream, post supercharger.

 

Also, think about the SR20 example. We're talking about a 1600+hp setup without an intercooler. They're running over 2k HP now, though with an intercooler and the engineer lead on it said himself that was a choice based on his experience when it comes to consistency, not because it was necessary... He also talks about how it's directly related to the pressure ratios seen on these smaller displacement engines, and that bigger displacement engines have the luxury of runner further right and lower on the compressor map.

 

Considering my target pressure ratios... I'll be fine.

 

As for BOV, that's one thing Garrett didn't include in the new G25 I'm disappointed about. Borg Warner included a REALLY trick recirculation valve in their EFR turbos that I think everyone should copy. Provides immediate built-in recirculation right at the compressor wheel which makes it ultra responsive while also making plumbing easy and sleek.

 

See here:

They also attach a boost solenoid for you too, so there's even less plumbing to worry about there as well. Overall, the EFR was designed for clean installs, and I like what they did about it. By contrast, if I use the Garrett I'll definitely wait for the wastegated hot sides to be available (currently in short supply as production ramps up) and I'll need to bring my own boost control solenoid along with BO/RecirculationV. 

 

...then there's also the chance that G25 dynos start coming out and are disappointing and don't live up to the hype... and then I'll just go EFR instead.

[seattlejester]

I agree, but based off what is seen in a traditional big block V8 with a roots type blower, they don't run post coolers. Engines that use remote superchargers that are roots like the jackson racing miata kit can run a roots type with a post cooler with fuel injection, but for the given V8 example people usually employ a supercharger and bolt carbs on top of them. My point is that those setups may work because of the carb pulling enough heat out of the inlet that the air has more capacity to be warmed by the super charger thus working. Throw the better cooling higher volume fuel of E85 and that works even more in their favor (Carb cooling + Fuel cooling)

 

If we look at some centrifugal setups that people run without an intercooler, but with a carb...taking out the carb would make those setups not possible without employing methanol injection or things of that nature 

 

For the example of the SR20, they are running M1 from a video I watched, looking that up it is 99.95% or purer methanol. Given the point above, that makes sense, if they are running almost pure methanol, the cooling effect would be huge. (Ridiculous fuel cooling)

 

Bottom line is I don't think those examples apply in this application. Turbo, no intercooler, fuel injection. Granted E85 does have a better capacity to cool given the higher volume usually used and ethanol content, but you would only have fuel cooling from E85.

 

^That above is all moot though, you can just throw more fuel at it like you say and run rich and keep the cylinders cool, you might not be operating in peak efficiency as you mentioned so all in all, probably not a problem that needs real addressing in any capacity unless you were looking for 10/10th of the efficiency. Just a fun example to contemplate and bench race.

 

The EFR turbos are pretty neat, but seeing those setups plumbed up I get a little bit annoyed with how the hoses have to route, the use of the plastic blow off valve kind of bothers me, granted I imagine they work fine.

[Gollum]

Well let's continue by discussion with some basic parameters of the topic.

 

This a great basis of data to work from, as it has many metrics in one spot: 

http://walshcarlines.com/pdf/fueltable.pdf

 

Ethanol is approx 2.64 times the cooling capacity of gasoline, volume for volume. But stochiometric is 61% lower, so you'll run 1.63 times more volume, effectively giving you 4.3 times the cooling capacity of gasoline.

Methanol is approx 3.71 times the cooling capacity of gasoline, volume for volume. But stochiometric is 44% lower, so you'll run 2.45 times more volume, effectively giving you 9.08 times to cooling capacity of gasoline.

(note that we didn't factor in weight difference which is what stochiometric is based on, but gasoline also has a weight range wider than the alcohols, so it's not as relevant)

 

This means that methanol is 2.11 times more effective than ethanol when comparing to gasoline. Now, we also have to account that we're comparing pure ethanol above, so there's also a 15% drop in all these comparisons when talking E85, which is more like E65-70 in the winter.

 

The SR20 engine is pushing 85+ psi..., and making 1000hp per liter... (on the known post intercooler changes this year, I can't find exact specs from last season, but they weren't MUCH slower... just less consistent by a good margin)

 

...I'm talking about running 25psi max, likely less than 20, and making.... 500ish hp? Maaaaaybe 600 if I'm dreaming after major headwork? So ballparking the math (we having fun yet?! I am  ), 1600hp divided by 2.11 would be 758HP. 85PSI would be 40PSI. Drop those numbers to 85%, and we're at 644hp and 34PSI. And well guess what... that looks REALLLY close to what I see on forums that people get on E85 with SR20s.... go figure 

 

The way I see it is that a really well-optimized L28ET should be able to reach about 200whp on pump gas without an intercooler. Plenty of people reach 180whp without an intercooler just fine. The largest problem is being within the compressor efficiency range, which doesn't take much to get outside of on the stock turbo. IF people built (and they don't) a turbo setup on the L28ET optimized for 250ish crank HP with a high-efficiency modern compressor, getting beyond the 200whp range without an intercooler might be quite possible or even common on 91 octane gasoline. Regardless of that though, my plan is to tune what I can on pump gas then tune E85, and let the ethanol sensor and map blending do its work. If I run out of ethanol, (I'm going to be dual fuel) my tune should be able to lower the boost and run on the gasoline-only map. If I'm running a nasty low ethanol content tank then it'll adjust accordingly. And if I get a good tank of E85, it should match my initial tune nicely.

 

Also keep in mind, by the time I get around to fabbing these, I'll also be doing a bit of head work, so I'll be at less PSI per HP compared to a bone stock F54/P90 combo. I'd likely be generating excessive heat at just 10psi on the stock turbo... But the target non-intercooled HP on pump gas mentioned above would still be reachable, if not easier so.

 

At the end of the day, I'm also not sure how much it REALLY matters to inject fuel before or after charging. There's a lot of heritage history with carbs on top of blowers, and blow through carbs were always more difficult to manage than draw through. The reason blow through was even a thing on turbocharged engines was that fuel is a lot harder on compressors than roots blowers, necessitating to not sling fuel through the turbo. One thing to keep in mind too, is that the LATER you add fuel, the FINER the mist needs to be. This has been a very real necessity with direct injection, as without a VERY fine mist you'll LOSE atomization capability over good port injection. It's also why with staged injection you're always going to run the later higher HP stage further up the airstream before the valve, to give more time for the fuel to atomize (and cool).

Edited April 27, 2018 by Gollum
typos, clarity

[seattlejester]

Oh I'm happy to discuss. Granted this seems to have shifted more towards feasibility rather efficiency. 

 

That is a pretty neat chart, but it doesn't include E85 and I'm not sure you can extrapolate for solution mixtures. For solids racemic mixtures really compromise characteristics, granted with a solution and the pure burn characteristics that may not apply. 

 

Also you really can't just divide numbers like that. While PSI is a pressure rating and you can have half the pressure, we are talking about a race engine with most likely un-comparable flow numbers. From what I could find a stock MN47 head flows 175cfm where as an SR20 flows over 200, the VE apparently flowing up to 280cfm modified people are quoting over 300cfm. I imagine an engine employing a billet block would have an even crazier head. Why does that matter? Well in a simple air pump situation if the head can only flow a certain amount the air that does not make it into the chamber will stay behind and build up which we read as PSI. So you could have a massive turbo feeding a really poorly flowing head and make 100hp at 90PSI, while on the other end of the spectrum you can have a very well flowing engine make over 900hp at 10PSI (I am employing hyperbole here). Granted with forced induction we have the more likely situation of you can turn up the boost and thus create more pressure in the system which would push more air into the cylinder so you could hit 300hp with a poor flowing head at 20 PSI vs the same engine with a good flowing head at 15PSI making 300hp.

 

Also if we are also just throwing estimated numbers out, the last numbers and the ones that mazworks quotes is 1800hp per 2.2L. That brings it down to 818hp/L.

 

Overall I think using that engine is really not comparable. We are talking crazy race engine that has to run maybe 30 seconds at a time a couple times a year. Employing race fuel and ram air etc etc etc.

 

The point being raised is not really the injection of fuel before or after charging, but fuel delivery. On the big block blown motors you have carbs that have fuel pulled down a passage way. You can see it if you look through the top it is almost a stream, as it is pulled it atomizes and vaporizes and cools the surrounding intake charge and such. You can sometimes find ice forming on the carb with how much heat it pulls from the surrounding. Injected motors don't really compare here as the injectors sit in the intake or in the cylinder in case of direct injection, the path is also usually much shorter, and it comes out atomized to an extent. Overall the mentioning of this is to point out you shouldn't be pulling examples from this camp (the big engine, super charged setups with carbs) as they don't apply as linearly to your proposed setup.

 

 

 

So the long winded point reiterated is, using supercharged big block engines with carbs or a ludicrous race drag engine are probably not places to pull numbers from.

 

 

 

 

With that aside the comparison that makes the most sense is in other L28ET setups. Preferably L28ET setups running E85, your desired turbo HX35 if I recall, and no intercooler. I'm not sure that exists.

 

We can look at similar engines in terms of displacement and turbo charging like 1jz and RB25, not sure anyone has deleted the intercooler in lieu of running just E85, not sure if that exists either, I did find this which seemed interesting, but his results seem less than stellar.

https://www.carthrottle.com/post/w8lxgbz/

 

The only way we would really know for sure is to put an air temp sensor after the injectors to see if the air charge is cool enough with E85 to be comparable, given where injectors sit that probably is not possible. Or I suppose a knock sensor would tell.

 

 

[ZHoob2004]

I haven't done any research on this, and I'm probably not thinking very well, but does it really matter where/how the fuel is added as far as the strength of the cooling effect? If we have the same mass of fuel, at the same temperature, vaporizing/burning, assuming the same EGT, shouldn't it have absorbed the exact same amount of energy?

[Gollum]

On 4/27/2018 at 3:04 PM, seattlejester said:

Oh I'm happy to discuss. Granted this seems to have shifted more towards feasibility rather efficiency. 

True, my last post was stabbing towards the "reality factor" not the "how much am I giving up factor".

 

Quote

That is a pretty neat chart, but it doesn't include E85 and I'm not sure you can extrapolate for solution mixtures. For solids racemic mixtures really compromise characteristics, granted with a solution and the pure burn characteristics that may not apply. 

Yes, E85 isn't listed, and not ALL fields can be calculated by mixing numbers, as many things have dynamic effects on the fuels. But some things like the latent heat of vaporization scale very linear in the blend. The evidence of this can be seen in how well certain aspects of flex-fuel tuning work all the way from E10 to E85 and anywhere in between with simple table extrapolation. While we can get stuck in the details of science, real-world tuning application says that most things blend linearly just fine.

 

Quote

 

Also you really can't just divide numbers like that. While PSI is a pressure rating and you can have half the pressure, we are talking about a race engine with most likely un-comparable flow numbers. From what I could find a stock MN47 head flows 175cfm where as an SR20 flows over 200, the VE apparently flowing up to 280cfm modified people are quoting over 300cfm. I imagine an engine employing a billet block would have an even crazier head. Why does that matter? Well in a simple air pump situation if the head can only flow a certain amount the air that does not make it into the chamber will stay behind and build up which we read as PSI. So you could have a massive turbo feeding a really poorly flowing head and make 100hp at 90PSI, while on the other end of the spectrum you can have a very well flowing engine make over 900hp at 10PSI (I am employing hyperbole here). Granted with forced induction we have the more likely situation of you can turn up the boost and thus create more pressure in the system which would push more air into the cylinder so you could hit 300hp with a poor flowing head at 20 PSI vs the same engine with a good flowing head at 15PSI making 300hp.

3

Well, as discussion basic bowl cleanup, mild unshrouding, and intake port centerline lifting will all be done. I'd be surprised if I'm not flowing over 190cfm on the intake side. But speaking of just dividing numbers, you also didn't note the difference between the number of intake runners when comparing here.  But anyway, the main compounder of heat in force induction is PSI, and the restriction of the air pump is directly related. I think we've both proven we understand that 100psi of boost might make 100 hp on some motors, and other motors might make 1,000hp with just 10psi. It's all about flow. That said, The G25 turbos should match stock-to-hot street L28 builds nicely, though I think it's a bit low on flow potential on the turbine side for a real hot L28 trying to chase world records. The compressor on the G25 I think is slightly biased towards higher PSI than I think most would really want to run on a L28 unless you're going to run low compression. This is also evident in the recommended displacement range Garrett advertises of 1.4-3.0 litres. An L28 is knocking on the door of that, but would likely be just fine as it's a mild flowing head. This is a turbo that WANTS to live around 2.5 pressure ratio and will work up to 3.5 in extreme cases. In my use case I'd be living between 2 and 2.5 pressure ratio, and where I land on the efficiency would depend on how well the head is flowing, which would also determine achieved power.

 

 

Quote

 

Also if we are also just throwing estimated numbers out, the last numbers and the ones that mazworks quotes is 1800hp per 2.2L. That brings it down to 818hp/L.

2

Good point, though they're dynoing over 2khp now , so there is that.

 

Quote

 

Overall I think using that engine is really not comparable. We are talking crazy race engine that has to run maybe 30 seconds at a time a couple times a year. Employing race fuel and ram air etc etc etc.

 

All true, and I agree and knew all this ahead of time. My point of including it in the discussion from the get-go is that they weren't running an intercooler, and were still running sub 7 second 1/4 miles and added an intercooler NOT because of heat management issues, but heat CONSISTENCY issues" Their goal with the new build was all about keeping times consistent. 

 

Quote

The point being raised is not really the injection of fuel before or after charging, but fuel delivery. On the big block blown motors you have carbs that have fuel pulled down a passage way. You can see it if you look through the top it is almost a stream, as it is pulled it atomizes and vaporizes and cools the surrounding intake charge and such. You can sometimes find ice forming on the carb with how much heat it pulls from the surrounding. Injected motors don't really compare here as the injectors sit in the intake or in the cylinder in case of direct injection, the path is also usually much shorter, and it comes out atomized to an extent. Overall the mentioning of this is to point out you shouldn't be pulling examples from this camp (the big engine, super charged setups with carbs) as they don't apply as linearly to your proposed setup.

3

Carburettors suffer from icing for a few reasons or might be more accurate to say that they have a few things working against them that leads to icing. And it should be noted that icing is an issue in LOW power setups, not just HIGH-performance application. First off, is that in order for carburettors to work properly they need a venturi, and the associated pressure drop from the venturi effect is actually the most fact proven reason for carb icing, and why low latent heat of vaporization gasoline can still have carb icing in 100 degree weather. Now, when you combine the venturi effect with poor atomization you end up with a very localized heat transfer as the fuel makes it's lazy path to vapor. 

 

 

Quote

So the long winded point reiterated is, using supercharged big block engines with carbs or a ludicrous race drag engine are probably not places to pull numbers from.

1

Likely a fair statement. Though most of the guys I've seen running without an intercooler with E85 and with fuel injection are still the V8 crowd. I think this is likely the fact that they run WIDE on compressor maps, with high flow and low-pressure ratio. These guys are making peak power around 6,000 in many cases, and have no issue spooling a large turbo by 2k rpm and don't mind turbo lag because of the off boost power available. There's no reason they couldn't run a setup with a higher pressure ratio, but the that starts to be a pretty serious street engine that few actually shoot for or attain. Also, a lot of the real world examples I've talked to of guys making 1,000+hp on methanol with blown small blocks and 12:1 compression aren't drag motors, they're in BOATS. And not drag boats. These are engines built to happily run cracked open producing 75% of its available power for an hour or more. I've never bothered to ask them how much they spend on fuel... But my point is that the engine builders in this market segment do this regularly and it's no big deal. They're running high compression, and no intercooler. And it's fine because of the relatively low pressure ratio. The fact they're running a carburetor should only be hurting them, as evidence that nearly all carb to efi conversions allow for more aggressive timing, which can only be attributed to fuel distribution being more even, along with in chamber charge temps being lower.

 

Quote

With that aside the comparison that makes the most sense is in other L28ET setups. Preferably L28ET setups running E85, your desired turbo HX35 if I recall, and no intercooler. I'm not sure that exists.

 

Yeah, I was looking at the HX35... back when I started this thread. How many years ago was that? I was also thinking I'd be doing all this on the cheap. Last year I spend 2k on an OEM Subaru turbo for my outback. The idea of spending that kind of money on a fancy aftermarket turbo isn't a huge deal in my mind anymore.  I'm also aware that I'm going to be in a fairly small group in the Z community running a setup like this.

 

Quote

We can look at similar engines in terms of displacement and turbo charging like 1jz and RB25, not sure anyone has deleted the intercooler in lieu of running just E85, not sure if that exists either, I did find this which seemed interesting, but his results seem less than stellar.

https://www.carthrottle.com/post/w8lxgbz/

1

IMO, that guys issues were likely 100% tune related. Guy even says he suspects his head gasket blew from timing. I think a better example is Perrin's article on the STI engine:

https://www.perrin.com/blog/post/your-sti-doesnt-need-an-intercooler-right

 

Personally, think their biggest issue was that they were still running gasoline in the port. It'd have been a different story with E85. I say that because I've seen an STI go from 375whp range and be VERY timing restricted (as in, they go from being down 30hp to knocking within a 1 degree spread) to being 450whp and nearly not caring about timing on E85. Was it intercooled? Yes. Was it low compression? Yes. But the timing map showed that for that power level, that 2.5 liter engine would have done just fine adjusting those variables, as it had timing to spare.

 

 

Quote

The only way we would really know for sure is to put an air temp sensor after the injectors to see if the air charge is cool enough with E85 to be comparable, given where injectors sit that probably is not possible. Or I suppose a knock sensor would tell.

 

This is something I've considered since I'll be injecting e85 far back near the plenum. A small thermistor style sensor would be pretty easy to poke through a small hole and seal up with minimal flow restriction. Also, knock sensor would be a very crude way to measure this. The real facny way is to install in cylinder pressure sensors, but that's beyond most of our budgets.

 

On 4/27/2018 at 3:46 PM, ZHoob2004 said:

I haven't done any research on this, and I'm probably not thinking very well, but does it really matter where/how the fuel is added as far as the strength of the cooling effect? If we have the same mass of fuel, at the same temperature, vaporizing/burning, assuming the same EGT, shouldn't it have absorbed the exact same amount of energy?

 

Yeah, I'm not sure it really does in the grand scheme of things. If all the fuel vaporizes, it has the same net effect. The struggle is that some engines have a hard time vaporizing all the fuel in certain load/rpm regions with carburettors. Why do you think there's always a mid range torque increase with EFI swaps on V8's? Usually a drop in peak HP is due to airflow restrictions, but that huge mid range change can't all be attributed to flow....

 

 

 

Now, let's look at some other maths since we're interested in known things we can actually look at and consider.

 

For example, for any given system we can calculate INLET AIR TEMP -> Compressor Efficiency -> PSI -> Outlet Air Temp

Assuming :

100 degree inlet air temp (either higher underhood temp or hot day ram air)

75% compressor efficiency (which I admit is high, but the G25 compressors reach 79 peak...)

targeting 20psi boost

at sea level pressure

Our turbo outlet temps will be a scorching 309 degrees.

 

Now, an intercooler that's able to achieve a crappy 50% efficiency would drop us all the way down to 203 degrees, which is "managable" and I've seen people tune around IAT temps in that range pretty often.

 

So the question becomes "how much does E85 drop temps"? This might be hard to answer definitively with data sheets on a paper, but some of these guys seem to have foumulas that match their real world experience:

http://www.modularfords.com/threads/194918-What-Is-The-In-Cylinder-IAT-Drop-With-e85

Specifically post number 11

 

Quote

It is easy to show that by assuming an isobaric process and constant latent heat of vaporization (Lv), that the drop in air temperature (ΔT) from the evaporation of the fuel is given by the simple expression,

ΔT = (Lv/cp,a)/AFR

where cp,a ~1 kJ/kg.°C is the constant-pressure specific heat of air, and AFR is, as you know, the air-to-fuel ratio. This is how I determined the numbers I gave to Eric B.

For E85, I calculate Lv = ~770 kJ/kg and AFRs = 9.8. (AFRs = stoich air-to-fuel ratio). So for an equivalence ratio of 1.25, (equiv ratio = 1/lambda), the change in charge temp would be,

ΔT = 770/(9.8/1.25) = 98.2 °C

To convert to a change in temperature in °F, multiply by 1.8, (but don’t add 32!), so 178 °F, which is what you got.

Read more at: http://www.modularfords.com/threads/194918-What-Is-The-In-Cylinder-IAT-Drop-With-e85

 

 

You can see they're just doing latent heat of vaporization by weigh compared to AFR. Though maybe simple it seems to be pretty effective and doesn't try to get into "real world modifier math modifiers" to compensate for lack of initial math function viability.

 

Now, the one thing I'd point out, is that using a lambda of 1.25 might be "ball park" for most people, but that's a bit on the lean side of your window to work with. The 1.25 figure is relating to the lambda equivalency ratio that you'd see on a chart like this:

 

 

Lean max torque for E85 lands at bout 1.15 while rich max torque lands all the way up at 1.40. That's a BIG range for "max torque" let alone "will burn without issue". I've seen E85 lambda readings down to .55 without ill effect on spark.

 

If we plug that kind of insane AFR into the calculation:

770/(9.8/1.8) = 141 (Celsius) * 1.8 = 254 (Fahrenheit)

 

That's 250+ degrees of temp drop capability. Now, some of that will "bleed" into intake and head temp cooling, since we're not using direction injection (would would still have some loss to head temp, but a lot less overall loss since there's no time spent in the intake and no pooling at the valve). All told, that's a massive amount of cooling capacity. But even if we look at this in the more realistic realm of .7 lambda we get this:

 

770/(9.8/1.4) = 110 (Celsius) * 1.8 = 198 (Fahrenheit)

 

We're still looking at enough cooling to match a crappy small intercooler, or even best it.

 

The obvious question though is: "If you're only going to drop the temp AS MUCH AS an intercooler, why do you think you can still run high compression if the net in chamber temp is the same?" Well, the answer to that is more to do with the way E85 burns than it does temperature. In my experience, E85 burns in a much more controlled fashion and much less explosive than gasoline. This allows you to approach minimum required timing and test max available timing for peak power in much safer and controlled way. Now, if you're stupid it'll still blow head gaskets at will since running too much timing for a given condition can still create insane pressure spikes, but if you're running on a steady state dyno doing timing traces you'll see the power drop off with timing in a much broader range before detonation becomes the limiting timing factor.

 

Overall, I expect to leave power on the table. I don't expect to win any dyno awards here. What I do expect, is to have a very responsive engine that starts up with ease and doesn't eat seals (since it'll run gasoline at start and shut down), while making more than enough power to be traction limited in almost all scenarios (trying to keep with either 225 or 245 max tire).

[seattlejester]

If the fuel vapor has a longer path to travel and is injected more in a stream like in a carb vs a spray as in an injector it can absorb more heat overall, I think through surface dynamics. Say you spray a mL of water vs drip a mL of water on the table. I think the overall heat absorbed may be the same for it to evaporate in a closed system, but the time the drip/puddle takes to dry would be longer the surface area of the droplets would expose it much more to air.. I imagine that has something to do with fuel efficiency as well as gollum states, the injected fuel is in a for lack of a better term in a more combustible state, where as the stream being dispersed through the Venturi might not quite be a pure mist by the time it gets to the intake port, that is where we have all the crazy edge of science things with grooved intake valves, and stuff that supposedly benefits carb'd motors because it allows one final spot for the spray to kind of tumble and aerosolize. Granted that kind of stuff is in the same camp as singh grooves so kind of pseudosciency I imagine and a digression.

 

I'm still reading through the response, but my goodness, level headed, thoughtful, kind of not expected over the internet, part of why I love this forum.

 

This is what I would really consider a discussion. I think we have slightly different views, but as a trial I don't think this will be problematic. With good monitoring or safeties in the ECU you would know before the motor started acting up.

 

This E85 no intercooler thing is a bit of a mystery, I imagine intercoolers are just kind of the turbo system, so removing it seems off to many me included. Then there is the compounded problem of measurements. Short of a dyno, our proposed post injector temp sensor it is hard to sort through examples found online. Some people state, "oh the boost went up when I removed the intercooler," well hotter air is less dense and could raise the system temp, using IGL we see that pressure would go up as well. Other people install an intercooler and note, "oh I lost peak boost psi with the intercooler," while you can get a pressure drop from the large volume, the intercooler also condenses the air by bringing the temps down so the air is cooler, more dense, and pushes less in the system bringing overall PSI down, but in this case making more power. 

 

I appreciate the discussion, I'm very curious to see the findings when this is made.

[Gollum]

46 minutes ago, seattlejester said:

If the fuel vapor has a longer path to travel and is injected more in a stream like in a carb vs a spray as in an injector it can absorb more heat overall, I think through surface dynamics. Say you spray a mL of water vs drip a mL of water on the table. I think the overall heat absorbed may be the same for it to evaporate in a closed system, but the time the drip/puddle takes to dry would be longer the surface area of the droplets would expose it much more to air.. I imagine that has something to do with fuel efficiency as well as gollum states, the injected fuel is in a for lack of a better term in a more combustible state, where as the stream being dispersed through the Venturi might not quite be a pure mist by the time it gets to the intake port, that is where we have all the crazy edge of science things with grooved intake valves, and stuff that supposedly benefits carb'd motors because it allows one final spot for the spray to kind of tumble and aerosolize. Granted that kind of stuff is in the same camp as singh grooves so kind of pseudosciency I imagine and a digression.

Only thing I'm going to bug you about in this post is that fuel doesn't burn if it doesn't vaporize. If fuel vaporizes it has created temperature drop by rule of physics. This means that we can measure "vaporization" efficiency with a smog dyno

 

Point: The vast majority of fuel gets vaporized, especially in port injection, and with much greater accuracy, hence efi running cleaner (when tuned).

 

I don't think I can over-stress though, that OEM's want to move to DI, and why? If 100% vaporization were harder with DI, I'm not sure they'd be so eager...

[ZHoob2004]

For clarity with my comment about heat absorbed by fuel, I'm assuming that all fuel is vaporized and/or burned due to chamber heat/pressure by the time it reaches the exhaust port. Essentially, more fuel should always lead to lower EGT (from which we infer chamber temperature), and where the fuel enters the system shouldn't really matter as far as heat absorption. At the same time, entry point and atomization do have effects on power and efficiency, where it is most often relevant.

 

Another angle on this: when running water/meth injection, does it really matter where in the intake tract it's being injected? My understanding was it's the phase change that's responsible for the cooling effect (AC for your combustion chambers), and there's plenty of heat/pressure change for this to occur.

 

Another, another angle: if it doesn't run cool enough there's always water/meth injection

Edited April 30, 2018 by ZHoob2004

[seattlejester]

58 minutes ago, Gollum said:

Only thing I'm going to bug you about in this post is that fuel doesn't burn if it doesn't vaporize. If fuel vaporizes it has created temperature drop by rule of physics. This means that we can measure "vaporization" efficiency with a smog dyno

 

Point: The vast majority of fuel gets vaporized, especially in port injection, and with much greater accuracy, hence efi running cleaner (when tuned).

 

I don't think I can over-stress though, that OEM's want to move to DI, and why? If 100% vaporization were harder with DI, I'm not sure they'd be so eager...

 

Ooh, excellent point. 

 

53 minutes ago, ZHoob2004 said:

For clarity with my comment about heat absorbed by fuel, I'm assuming that all fuel is vaporized and/or burned due to chamber heat/pressure by the time it reaches the exhaust port. Essentially, more fuel should always lead to lower EGT (from which we infer chamber temperature), and where the fuel enters the system shouldn't really matter as far as heat absorption. At the same time, entry point and atomization do have effects on power and efficiency, where it is most often relevant.

 

Another angle on this: when running water/meth injection, does it really matter where in the intake tract it's being injected? My understanding was it's the phase change that's responsible for the cooling effect (AC for your combustion chambers), and there's plenty of heat/pressure change for this to occur.

 

Another, another angle: if it doesn't run cool enough there's always water/meth injection

 

Assuming the same amount of fuel in a closed system assuming perfect atomization regardless of method, then the heat absorbed via vaporization should be equivalent.

 

Carbs though as mentioned have poor atomization compared to fuel injection, and as a byproduct probably and in actuality do require more fuel to have the equivalent fuel to keep the AFR happy. That combined with larger size droplets means less surface area if my recollection is correct means slower vaporization.

 

Disclaimer of course, this is pure bench racing for me, I've only dealt with this stuff in theory. 

 

Some meth injection systems employ almost a hose, while others employ a spray nozzle, so there must be some disagreement there. Similarly they do offer almost direct port methanol injection now where it used to be a single pre throttle body injection used to be the norm. 

 

The guide I looked up said with direct port methanol injection you can actually remove fuel, I imagine more of the methanol actually makes it to the cylinder and acts as fuel. Where as injecting it further upstream it would serve more as a method to cool the charge and not as a fuel additive?

 

Not really sure on that, but yes meth injection would be another method of cooling the charge and enriching the fuel, granted if he is already planning on E85, just adding a bit more fuel to be used as a sacrificial heat sink would be simpler. 

 

I just thought of this though, adding too much fuel can result in washing the cylinders, given how E85 behaves and how it usually ends up in the crank case, maybe the above would not be as recommended. 

[Gollum]

Well, and to throw a noodle into the bowl of fruit loops, there's a good number of people that are using straight water injection pre-turbo. With a fine enough nozzle and enough volume the aim here is to not change the compressor pressure ratio, but to improve output via basic thermodynamics. Nowhere on a compressor map is it assumed that compressor inlet is a perfect atmospheric barometric pressure. Lower the inlet pressure via temp drop via water, and suddenly you've increased max compressor flow. Great way to cheat a small turbo into flowing more air.

 

Personally, I think it's a neat idea but has limited use cases. If your turbo isn't running off it's map already, then you're just adding another molecule that's robbing space that could be used by oxygen. Though, if I DO end up temperature limited in my build, it might be a viable option...

[Gollum]

2 hours ago, seattlejester said:

I just thought of this though, adding too much fuel can result in washing the cylinders, given how E85 behaves and how it usually ends up in the crank case, maybe the above would not be as recommended. 

 

The largest concern I have is with shutdown. I've seen plenty of people run e85 daily for long periods with little effect on wear. What's more amazing is how clean everything is at teardown.

 

My larger concern is rust on cylinder walls if you don't drive for longer periods (like over the winter, because fair weather car). This is one of the reasons for going dual fuel. Running pump gas off boost before shutdown should allow the cylinder walls to stay coated between running.

Edited April 30, 2018 by Gollum

[seattlejester]

Hmm, I've heard the unspoken side of E85 is the 1k recommended oil change interval, perhaps a byproduct of people running too much fuel to eek more out of their setup. The duel fuel thing sounds pretty neat, I imagine secondary injectors or something of that nature and a switch over to an auxiliary tank? Then again that seems pretty complicated in its own way.

[Gollum]

It's actually not that hard these days, even in megasquirt:

 

http://www.msextra.com/doc/pdf/html/Megasquirt3_TunerStudio_MS_Lite_Reference-1.5.pdf/Megasquirt3_TunerStudio_MS_Lite_Reference-1.5-216.html

 

You even get to adjust the range/curve of the switch method you use:

http://www.msextra.com/doc/pdf/html/Megasquirt3_TunerStudio_MS_Lite_Reference-1.5.pdf/Megasquirt3_TunerStudio_MS_Lite_Reference-1.5-221.html

 

You can also adjust target boost curve on each table, so if for some reason the switch never happens you can have a sane boost target for whichever map you're on. In flex fuel mode using the E85 sensor this would be handy in case for some reason your ethanol station filled the tanks with regular gasoline (it's happened, and blown up motors):

http://www.msextra.com/doc/pdf/html/Megasquirt3_TunerStudio_MS_Lite_Reference-1.5.pdf/Megasquirt3_TunerStudio_MS_Lite_Reference-1.5-225.html

 

 

In short, you tune with pump gas without the switch over. You set a tunable boost curve and spark limit the motor so that it stays safe if boost runs away beyond fuel capability. You then move on to tuning with JUST the secondary fuel (E85 in my case) and get it tuned, with particular attention to where you'll be using it most, under boost. Once that's done the flex fuel sensor will do it's work to blend between the two extremes based upon sensor input. This whole process is done on your E85 injectors.

 

Now, in my case you likely already have a working pump gas tune on another set of injectors. And that's where this comes in:

http://www.msextra.com/doc/pdf/html/Megasquirt3_TunerStudio_MS_Lite_Reference-1.5.pdf/Megasquirt3_TunerStudio_MS_Lite_Reference-1.5-116.html

 

This allows a injection switch over from one set of tables to a second set of tables, along with precalculated adjustments for injector size differences. Obviously your small injectors might need some trimming to get back to perfect on pump gas, but odds are it'll be darn close, possibly just needing some overall trimming, not detailed cell fixes.

 

That said, in this mode, your essentially using TWO tables (each) for fuel and spark. MS3x supports FOUR tables, which means there's a way to configure that I haven't figured out yet in which in theory you should be able to get all four potential scenarios working (flex sensor on both tanks with a table for each plus staging switch over)

 

As for the oil changing every 1k. I've heard people say that, and I think it's paranoia. I've met people that tune E85 for a primary living, and they suggest typical 3k rotations on street driving, or every race event if you're tracking it regularly, but they recommend that for pump gas too.... I think the harsher reality is that going from pump gas to E85 when tuned properly is AMAZINGLY fun as the car behaves so differently, so you spend a lot more time with your foot in it. And anything driven that hard should have it's oil changed regularly.

 

[seattlejester]

Oh I had no doubts about the electronics involved, more of the physical aspect as in two fuel pumps, two fuel rails, two fuel lines, etc etc. At that a same sided intercooler seems perhaps an easier task.

[Gollum]

13 minutes ago, seattlejester said:

Oh I had no doubts about the electronics involved, more of the physical aspect as in two fuel pumps, two fuel rails, two fuel lines, etc etc. At that a same sided intercooler seems perhaps an easier task.

I blame 510six for starting me down a rabbit hole of research about 7 years ago. He was also running dual fuel staged injection with E85 on the secondaries and pump gas on the primaries. He also was running no intercooler. He also put 400 to the wheels... ...off the bottle...

 

...What shocked me most was how LITTLE E85 he'd burn through. He had a small 3 or 5 gallon tank that he wasn't really ever worried about disappearing without just cause. Could have a lot of fun with minimal impact on the "fun tank".

 

The point in my mind isn't just avoiding an intercooler, that's a byproduct. The goal is high mpg on cheap pump gas, and also being able to use "cheap" race gas (e85) without bothering with the drawbacks of needing to go get the good stuff when you want to have fun. But running e85 in cruise condition on the highway mile after mile isn't necessary in any measurable way. Sure it's more work to dual pump dual tank, dual injector, etc. But the benefits are great in the long run.

 

Though his (510six) sanity shouldn't be what we consider "benchmark", and this is closer to his car's current evolution: