pzr2

iTrader: (3)

Hey everyone, (again, haha)

I've been obsessed with the idea of a separate reservoir for premix since I started premixing. I've always wanted a big tank of it in the spare tire cavity, which constantly mixed the right amount into the fuel (and/or to an OMP adapter) on demand. Directly pumping the premix from the reservoir into the gas tank never worked out in my head though, because how would you know how much to pump in and what if the tank was only topped up partially?

I have plans of installing a surge tank in the car now, so I can probably get around all of that. I'm going to be running a standalone ECU and programming a couple of microcomputers to handle some other functions in the car, so it wouldn't be a stretch to take all of the telemetry I have of the fuel system to calculate the outgoing fuel and match the proportional amount of premix to pump into the surge tank. So that's mostly taken care of.

My question is, given the amount of turbulence that occurs in a surge tank, would it be reasonable to expect that in exchange for the small amount of time the premix would spend mixing with the fuel, pumping premix into the surge tank would mix sufficiently?

Thanks,
Luke

WANKfactor

iTrader: (1)

It would mix its absolute **** off. Only problem - youve got fuel returning from the rails back to the surge tank and fuel returning from the surge tank to the fuel tank at full capacity the whole time. It would be pretty much the same as putting oil directly into the fuel tank because it all just circulates. You cant dispense oil in real time from there like an omp does, unless i'm missing something?

pzr2

iTrader: (3)

Oh, sweet, just what I wanted to hear. I've been trying to work this out in another thread, but I'm planning on converting over to returnless. Still not sure exactly how, but even if I wasn't, I could try passing off the injector pulse data from the standalone EMS to a separate microcontroller that would take the raw pulse data and rail pressure to calculate an outgoing flow from it. If you draw your control volume as the fuel system from the surge tank to the injectors, the only outgoing fuel mass is through the injectors and the only coming in is dumped into the surge tank by the lift pump. The mass flows should theoretically be matched if the system is in a steady state, because the return flow is already mixed with premix. The only problem I see to this is that there will still end up being a discrepancy between outgoing flow through the injectors and the inbound flow from the pump in real time (injectors: entirely load dependent, pump: entirely dependent on a steady flow speed unless the surge tank is at capacity, so they're dependent, but not equal to each other. Not a huge deal if lift pump has sufficient overhead to outflow the injectors so that there's negligible latency from out to in). I'm sure there's a bunch of different other ways to approach this (read flow rate inbound from pump using some sort of sensor), but the mixing is all I really needed to hear. Thanks!

EDIT: Also, I was under the impression that most people just return the fuel directly to the surge tank instead of back to the tank

WANKfactor

iTrader: (1)

^ conventional setup has return to surge, and overflow from surge back to tank. Interesting idea though.

j9fd3s

iTrader: (3)

here is a crazy idea.

why not run an oil injector, and just let it spray into the intake like a fuel injector?

TomU

iTrader: (11)

A lot of 2 stroke engines employ oil injection. I had an outboard 150 that I blew because the oil injection failed.

KISS. Just dump some oil in the gas tank

peejay

You could put a small mechanical positive-displacement pump on the engine to do this, too. It would automatically inject a certain amount per revolution. You might want to have some sort of control on it to vary the actual amount based on throttle position, or if you wanted to get fancy you could make that part electrically controlled.

WANKfactor

iTrader: (1)

Is there a way of beefing up a stock OMP system and then just running a rotary aviation set up?

pzr2

iTrader: (3)

Right, but you can also look at it as complicating the car's tasks to make the user's job more simple. You and I know a lot of times, it's the user that's far more unreliable.

Also, I probably mentioned it before, but I'm running a Megasquirt on a 13B-REW. I'm not aware of whether someone's managed to control the stock electronic MOP system with a standalone, RA adapter or otherwise. Moreover, it would still be injecting right into the combustion chamber, which would mostly just end up lubricating the center of the apex seal. If I was still working on the old S4 13B-T, I'd have just plugged the ports on the block and run the stock mechanical MOP with the factory ports on the LIM, but again, no 13B-T, and no intake injectors on the REW.

I might have to rethink things if there's going to be a return from the surge tank to the fuel tank.

RGHTBrainDesign

You do realize the mixture ratio of oil and fuel would constantly change with this setup, right?

If you used an Oil Metering Pump from your trunk reservoir and dumped it into the fuel as you used said fuel, that ratio would change once the return fuel (now already carrying oil) came back into the tank.

Solution:

Factory OMP + RA 2-Stroke Adapter
2-Stroke Reservoir in Engine Bay with a Low-Level Indicator
Premix 200:1 to 100:1 in your fuel tank and call it a day.

With e85, make sure your premix is alcohol stable. Grab some e85 and premix and test it yourself with a mason jar.

Peejay is the best troll. <3

pzr2

iTrader: (3)

Oh, heaven's sake. For the last time, I'm running a Megasquirt and have no control over the factory REW MOP.

I understand that there's an issue involved in ratio of premix present when there's a return involved. If you actually read what I was writing above, you'd have noticed I'm planning on going return-less anyways. Personally, my solution would have been thus if I wasn't going return-less:

-Install low pressure fuel regulator bypass within the tank to limit overfilling the surge tank and not overpressurize the lift system (flow to surge tank is now limited based on pressure and has no need for a return to the tank)
-Flow sensor anywhere between the in-tank regulator to the inlet of the surge tank to determine non-lubricated fuel flow into lubricated fuel system
-AEM-style water/meth injection system with bypassed constant speed pump to inject appropriate amount of premix into surge tank based on flow sensor data for a fixed lubrication ratio
-Eliminate return from surge tank to fuel tank
-Keep return from rails back to surge tank if you so insist

Moreover, I still have my reservations about spraying the premix directly into the combustion chamber. I personally don't believe that's enough to propagate the lubricant to the sides of the apex seals, side/corner seals, etc. I know it was good enough for Mazda, but I'm planning on making 2.5 times the factory power output.

And while my fuel system is E85 ready, I'm sticking to 91 and water injection. Thanks for replying anyhow.

RGHTBrainDesign

When you're premixing your oil into your fuel, your combustible energy by mass is lower, is it not?

How do you know that premix oil won't clog the 7-micron injector filters (yes, that's modern injector "strainer" size)? That one piece of knowledge alone goes to show that most "motorsports" fuel filters are absolutely inadequate to maintaining a reliable car.

"spraying the premix directly into the combustion chamber" - Well, this is a port-fed motor and you're essentially doing the same thing with the OMP vs. injecting it into the intake runner ports (which have zero restriction to the engine ports), except for now your fuel mass is consistent and contains more energy. Premixing at the fuel tank for a performance setup is counter intuitive for this very reason. Makes sense for added reliability, given that you're running a 3-5 micron fuel filter.

Why not use the same controller (or add another module) that's doing your AI and control the OMP with that? AI is darn near the least reliable system you could put on your car. I'm one who's against "spray" power, whether it be NO2 or Water or Methanol.




All of what I typed above doesn't matter. #1 It's your car and your money. Do what pleases you and don't be afraid to experiment. #2 I'm an ******* and an engineer. I see any weak points and I dump the entire idea; set in my ways until new information surfaces. #3 This is all for fun. We drive for fun, build for fun, etc. This forum is just impeding your adventurous side. Go out and try it all. Oil tank in trunk, returnless, etc. **** whoever says you CAN'T do something...

j9fd3s

iTrader: (3)

the metering pump is just a stepper motor, like an electronic throttle. there is also a position sensor, so the stock ECU can detect if the pump is working.

b. if the REW method of oiling the housing wasn't better, they wouldn't have used it. remember they want the same or less wear with less oil.
c. seems like the simple way would be to just run the stock metering pump, and throw some premix in the gas when you want to beat on it harder than normal

a, since we need to run some kind of seal lubricating oil anyways how does that matter, and B how much does the premix effect the fuel?

peejay

Most likely not. Arguably it is higher since the oil is denser per volume than gasoline, and as a general rule of thumb, all petroleum hydrocarbons are so close to the same BTU/lb that for back of the envelope rough calculation purposes, they're the same.

Practically speaking, even when premixing at high-for-rotary ratios, the amount of displaced gasoline is smaller than normal cycle-to-cycle fuel requirement variations. And even then, if you're tuning the car yourself, it would be normalized out anyway.

(Anecdote: We heat our building at work with a waste oil furnace. It can burn up to 25% gasoline by volume. It can also burn straight HHO or kerosene if we wanted, but we'd have to readjust the pump speed because of the different viscosity. Doing the math between HHO, non-road kerosene, and natural gas, they all come out to about a wash because they all have roughly the same dollar/BTU ratio. Waste oil is effectively free for us, so we use that )

You're... kidding, right? How large do you think an oil MOLECULE is? They could be .7 micron and it'd be fine.

peejay

Just for giggles and grins, I did a search.

Gasoline is about 125,000btu/gallon. Diesel is about 135-138,000btu/gallon. Common two stroke oils are also about 138,000btu/gallon.

So premixing actually, technically, adds to the energy of the fuel you are injecting.

Now, interestingly, I've seen many references to the "80 octane" fuel that Mazda used to run in IMSA. It was then-current race fuel denatured with kerosene. So the mix would have the ignitability of a highly volatile race fuel combined with the burn characteristics of a heavy fuel.

Interesting. I wonder how much of this was dictated by the ignitions systems they had to run back then. Running through a distributor leaves you very little time to charge a coil at 10000rpm.

Also, my searching for oil BTUs has me finding out that some truck Diesel enthusiasts swear by premixing 2-stroke oil in their #2. Well, there goes my idea for running a rotary on pure kerosene without premixing I still have crazy ideas for using a TDI electronic injection pump to make a direct injected rotary, too...

j9fd3s

iTrader: (3)

the GTU/GTO cars were at least running CDI, but yeah they probably were ignition limited.

Alfa Romeo used a diesel injection pump in their gas cars, there is something about oiling them that sort of rings a bell, but point is that it works well. Alfa also had VVT in 1981...

so put that TDI pump on your engine, i'm sure it would work fine

peejay

You're probably thinking of the SPICA fuel injection, which was a lot like a Diesel injection pump. Also very similar to the Kugelfischer mechanical injection used on the BMW 2002tii and Turbo.

I've never seen one in person, but they're supposed to actually have four little pistons in them!

Of course, modern Diesels are common rail with electric solenoid injectors ike EFI gasoline engines, and EFI gasoline engines inject the fuel directly into the chamber using Diesel technology... The future is weird.

WANKfactor

iTrader: (1)

dont worry about it. just get a couple washing machine motors and a few thousand mobile phone batteries and boom! future proof.