The Mega Supercharger 4AGZE thread, using twinscrew charger

[jondee86]

good tuner helps things stay together.
15psi is nothing for a 4agz

Yeah... I know that And I could pay someone to build my engine, fabricate the charger mounting plate, design and weld the ducting, install it, wire it up and tune it. But then I don't learn anything.

The whole reason for putting a supercharger on the car is to give me the chance to learn some new stuff and ultimately have the satisfaction of knowing that it was "built, not bought". And the reason I am on this forum asking noob questions is to get some input from people who have already played with superchargers. The companies that market supercharger kits are long on marketing BS and short on technical detail. Getting decent selection and application detail is a case of opening a crapload of oysters, and only finding a few pearls

Hopefully, with a bit of informed advice, I can put together a system that works, while avoiding making any big mistakes in design or construction. Up until this time the interwebz have not been able to find me anyone who has installed a twinscrew charger in an AE86. That means I don't really have anything to follow other than mr2swift who is wrestling with fitting one to an AW11 in a totally different configuration.

And you are right about a good tuner. Once the car is driving and everything is working as it should, it will go to a specialist to have the tune optimised.

Cheers... jondee86

[JustenGT8]

The car is not a daily... just a toy for fine days I don't really have the space to fit an intercooler, so that's why I was planning a progressive water/meth (or maybe just water only) with injection starting at (say) 6psi and ramping up with boost. That way I was hoping to only be misting when I was having fun... not during normal round town running.

How much does water/meth or straight water mess with the AFR ? What kind of affect does it have on the AFR ? Is it predictable, to the point where it can be compensated for in the fuel map ?

This is the kind of information I need for firming up the system design.

Cheers... jondee86

PS: I read that meth is corrosive to metal components, so for my modest requirements water only (ala Aquamist) might be the better option.

I only used water/eth (I used ethanol not methanol as much less hassle) on my roots setup. Mine wasn't progressive, just on/off with two nozzles ore throttle body and one in the inlet post SC. So a little bit complex and a little bit simple

WI can be quite a complex beast if you are using it to enhance torque. If just for intercooling then can be much simpler and any other gains you get are cream.

My issue was controlling det, so I did the calcs to try and get a 10-15% mix of water/eth around the rpm of peak torque.

It still seemed to keep intake temps under control above and below the ideal mix.

Numbers...17psi, 450cc/min WI for around 300rwkw with 55deg intake temps (intake temps were more than double that without the WI) from a M122 supercharger. This wasn't on a 4A obviously but the principles are the same. Oh, and on E85.... You'd be mad to run a non intercooled SC engine on 98 in my view.

[jondee86]

Hey thanks for that Justen. I have just been researching the water injection question, and at this stage it looks like starting off with straight water might be a good idea. Apparently water has virtually no effect on AFR's, but once you start adding alcohol the AFR's will go rich. And as a twinscrew charger does not raise discharge temps as much as Rootes type, adding water only may be all that is needed for det suppression.

I have a lot to learn about running an engine on boost (never did it before), so I intend to start out keeping the setup safe and simple. Once I have the basics sorted I will gear up for a bit more boost, and at that stage look at adding alcohol. And it seems that ethanol is the way to go... not really so keen on E85 as it is not a big thing here yet.

Again, for simplicity, I am leaning towards a single nozzle misting into the charger inlet once boost goes above 6psi. It would be nice to be able to ramp the flow rate up according to boost, and that's about where I am up to now

Cheers... jondee86

[jondee86]

Digging into the water injection options, I see that Aquamist use a Fast Acting Valve driven by a fancy PWM controller in their high end systems.

;;

And my previous investigation indicates that I should be able to create a PWM output from the e420c that follows the injector duty cycle. So the easy solution would be to get the valve and other essential components from Aquamist and take the control signals from the ECU. Size the nozzle for 15% of maximum fuel flow, start injecting water at (say) 6psi and then track the fuel injectors.

The idea is to match the water injection to to the fuel injection at all combinations of boost and rpm, rather than just switching on a fixed injection rate above a certain boost level.

Cheers... jondee86

[Duk]

Digging into the water injection options, I see that Aquamist use a Fast Acting Valve driven by a fancy PWM controller in their high end systems.

And my previous investigation indicates that I should be able to create a PWM output from the e420c that follows the injector duty cycle. So the easy solution would be to get the valve and other essential components from Aquamist and take the control signals from the ECU. Size the nozzle for 15% of maximum fuel flow, start injecting water at (say) 6psi and then track the fuel injectors.

The idea is to match the water injection to to the fuel injection at all combinations of boost and rpm, rather than just switching on a fixed injection rate above a certain boost level.

Cheers... jondee86

While I've been a slack **** for way too long(!), I did get a couple of FJO water injection solenoid valves for my WI intentions (2 different cars).
They're designed to be able to perform variable flow rate delivery, but playing with 1 of them years ago did seem to show a fairly non-linear flow behavior.
On top of that, they had a fairly high base duty cycle before they would actually flow any water with decent atomization. That is, you couldn't just ask for say 10% duty cycle and get a nicely atomized spray of water. This was done using an output from my Adaptronic and I did play with different frequencies to see if that helped.
BUT(!!!) I was using the typical high pressure type pump and mine had a pressure switch rather than the better internal bypass. That could have been causing some of the problems at low injector flow (low duty cycles) as the pump would trigger its over pressure switch, stop the pump, only to have the pressure drop and start the pump again. A bypass type pressure regulator would have been a much better pump choice.
These pressure switched types of pumps are probably much less of a problem with the systems that vary the pump voltage to effect nozzle flow, but logic says that if you change pump behavior to effect nozzle flow, then atomizing behavior of the liquid at the nozzle, will change too.
Maybe it's a problem and maybe it's not.

If you do go the route of the Aquamist HSV, it'd be a good idea to run the pump/valve/nozzle assembly off of the engine and get a bit of a feel for how it actually performs and behaves. And use either a bypass type pump, or a motor speed controller that slightly reduces pump speed at lower WI duty cycles.

[jondee86]

Good comment Duk I will have a fair amount of experimentation to do with the e420c. At the moment I have a theoretical setup using the second fuel map to adjust the water injection volume according to engine rpm and MAP. I'll need to revisit the theory as it has been a while since I last looked at this aspect.

The Aquamist system uses a 160psi pump with internal bypass, so the HSV always has 160psi, and the pump capacity is way more than I will need. I haven't looked into the other hardware, but Aquamist have a nozzle with check valve, and if I use that with injection starting from 6psi boost at (say) 20% duty cycle, I would hope to avoid any "dribbling". I don't think that varying pump speed/pressure to control nozzle flow is a good idea. PWM is the way to go, so long as it is not as noisy as this old video https://www.youtube.com/watch?v=_HQMTfybwaA

Over the last couple of weeks I have looked at all the US based water/alcohol injection kits and none of them particulary impress me, They are big on flashing lights to dress up low tech systems as something special. Aquamist is a step above in my book, and they were using a HSV before anyone else, so although it is a bit spendy, I will most likely use their valve, pump, nozzle and some of the failsafe stuff.

I'm going to setup the e420c and just drive a small relay as an experiment. Hopefully I can at least get that to buzz at a rate that mirrors the fuel injector duty cycle.

Cheers... jondee86

[JustenGT8]

Yeah aquamist are the top of the heap by a mile....not that a simple system won't still work well. I have only used aquamist nozzles as they come in a great range of sizes and appear to atomise better than anything else I've tested.

[knightrous]

https://www.rbracing-rsr.com/waterinjection.html

800-1000psi pressure pump.

All of the WI kits I've looked at recently cost nearly as much as a W2A setup
So I'm just going to go W2A for my rig (1MZFE + M90)

[jondee86]

That RSR site is a damn good read Their philosophy about water injection systems is great if you are building a Bonneville bike or any other engine that is designed to push maximum power only within a narrow rpm range. I love the KISS principle, and a high pressure pump with a nozzle will certainly give superb atomisation. The downside is lack of flexibility when the engine is not operating under WOT in a narrow range. And this makes the RSR system less suitable for non-racing applications.

Once installed, an intercooler does not require further attention, so it is a good option if you have the space. According to what I read, water injection is supposed to offer other benefits beyond just cooling the charge, and perhaps that makes up for the hassles of refilling and monitoring against injection failure when you have the pedal nailed to the floor in the overtaking lane

Cheers... jondee86

[JustenGT8]

I actually have most of an Aquamist kit if you're interested? Pump, loom, pressure switch...no nozzles though

[jondee86]

I actually have most of an Aquamist kit if you're interested? Pump, loom, pressure switch...no nozzles though

Could be interested in the pump if it is the current 160psi internal bypass version ? The HSV, nozzle/s and any other bits should be cheap enough to get mailed from the UK, but the pump would probably be expensive to ship. Haven't really got round to thinking about fail-safe monitoring yet.

Cheers... jondee86

[Duk]

https://www.rbracing-rsr.com/waterinjection.html

800-1000psi pressure pump.

All of the WI kits I've looked at recently cost nearly as much as a W2A setup
So I'm just going to go W2A for my rig (1MZFE + M90)

The RSR site has some excellent info about water injection.
But me thinks they were blowing their own trumpet a little bit too much.......................

Check these articles out from Autospeed. In particular, notice the water pump.

http://www.autospeed.com/cms/article.html?&A=110368

http://www.autospeed.com/cms/A_113191/article.html

I've got 1 of those pumps and they do work a treat with that same nozzle. I was using it to improve the efficiency of my split system A/C. Exhaust temperature of the outdoor unit dropped down to ambient (44*C) at the time from 56*C. Discharge temp from the indoor unit dropped 12-13*C too. This was measured with a K type thermo-couple on my multimeter.

[jondee86]

Check these articles out from Autospeed. In particular, notice the water pump.

Interesting stuff !! Probably a little bit too much DIY for me. Getting this project together is going to be a big enough exercise without adding more unknowns. Now that the American pumps with internal bypass put out a reasonably stable constant 150-160psi, I am a lot happier to use one of them. Some of them are rated higher, only the Aquamist HSV is pressure limited to 150psi max, so I have to be content with that.

Having looked at a number of video clips of nozzles spraying in the open air, it appears that the majority of the spray would be hitting the intake ducting walls ? Or does the high velocity intake airflow whip the spray away before it hits the walls ? At some stage I guess I need to get a handle on that, so I can figure the best position to locate the (single) pre-charger nozzle.

EDIT: Probably answer my own question here If there was a problem with water with the water spray "condensing" on the inside of the ducting, presumeably adding alcohol to the mix (and reducing the water content) would get round the issue by making the spray more volatile. It would then flash to vapor rather than collecting as liquid in the intake. Possible

Cheers... jondee86

[Duk]

Really speaking, the best solution to adding water to the air stream would be done with a nozzle located centrally in the pipe and (obviously) spraying in the direction of the air's travel. The potential problem with that is, the further the solenoid valve is from the nozzle, the more lethargic the ON-OFF PWM nature of the valve becomes. And so the greater the tendancy for the nozzle to dribble (rather than atomize) at the begining and end of each pulse.
The pipe would also deliberately have a fairly corse internal surface finish to make sure the boundry layer of air that ran along it, constantly moved and tumbled to stop the water from sticking to it.

Something that really should be mentioned, but typically gets left out..................... Water injection's biggest heat absorption happens inside the combustion chambers!
The amount of time that the water mist spends inside the incoming, relatively cool air charge, is small. But the amount of heat that is present inside of each cylinder, especially around the exhaust valve(s) during the compression stroke, is HEAPS(!!!) higher! It's here where the water turns to steam and absorbs lots of heat energy. The true end results will be shown in lower exhaust gas temperature numbers, for a give ignition timing and AFR number, rather than IAT numbers.

[JustenGT8]

It's one of the earlier barrel pumps....120psi ish from memory, I'll dig it out.