DIY: Nitrous oxide setup for ~$100

[Daox]

3D print new MAF housing ftw.

You can place your injection point in your favorite spot.

Because you're a 3D print wizard.

I'm somewhat surprised I'm actually encouraging this...

That is what I was thinking. I'll need to get a aftermarket filter of some sort though.

Alternatively, I can mod the extra intake airbox housing I bought when I designed the original MAF housings. It wouldn't be too hard to punch a hole in it and stick the fitting in there. I'm just not sure its thick enough to hold on and maintain a good seal.

[Daox]

Modifying the solenoid valve - Part 2

Alright, its been mostly talk thus far. Good talk, but talk. Its now time to get down to business and do some of this work we've been talking about. So, lets get to modding the solenoid valve!

Here we have the valve body. The hole in the center is 2.5mm in diameter, and we need to make it .9mm in diameter.





To do this, we are going to use this handy 3/32" rod. 3/32" is 2.38mm, so it fits very nicely into the 2.5mm hole with not much slop at all.







I stuck the rod in the valve body, and made a mark so I could cut it off to the correct length.









Test fit, it sticks up just a little bit. This is perfect as I'll sand it to bring it down to the correct height.





Next up, time to drill the .9mm hole in it. For this and the jets (I'll get to those later), you will need a micro drill bit set. Below is the one I used. It was $13 on amazon.





First step, center punch the brass rod to help center the drill.





Next, I started drilling by hand. This went really slow, so I set it up in my drill press. You need to be VERY careful dealing with these small drill bits. They break extremely easily.







Now, the drill bits aren't that long. So, I notched out the brass rod with a dremel to intersect the drilled hole.





With the opening made in the rod, I sanded the rod down to the correct length. I made it close to flush with the old sealing surface.





Next, I sanded a chamfer on the brass rod. This will reduce the sealing area, and increase the pressure rating of the valve a bit more. Really, I should have calculated things with the sealing area, not the orifice diameter. I don't think it'll be a problem, but that is the more correct way to go.







Now, we don't want the valve's sealing surface to get the the way. So, time to grind it off. I actually went a bit farther and put a tiny bit of divot into the valve. The next step is soldering things, the divot will help hold the solder and keep solder out of the .9mm hole.











Alright, we're to the last step of the solenoid modification, soldering up the brass tube into the valve body. To do this, I used normal electrical solder, and also plumbing flux like you use for sweating copper pipes. This may seem like a weak material choice, but it will work fine. The brass rod is taking the load that is pushing down on it because it sits all the way down into the valve body. The solder is just holding it in place sideways and sealing the tiny bit of space around the rod between it and the valve body. Here is the stuff I used.

I used the flux from this kit.





I was going to use the plumbing solder until I looked at the diameter of the solder... The plumbing solder was just too large to use on the tiny solenoid valve. I was concerned with getting solder inside the .9mm hole.





I also used map gas. I would recommend propane, but I didn't have any. Map burns a decent amount hotter, and honestly too warm for this. So, I just had to be a little careful getting things too hot.





To prepare the parts for soldering, I did sand down the rod, and also took a tiny dremel bit to the inside of the valve body hole. This should make the joint a bit stronger. Then I coated the rod in flux, and stuck it in the valve body.

The actual soldering process is really easy. Just warm up the part for a while, take the flame away, and touch the solder to the joint. If it melts and sucks into the gap, its warm enough. If it doesn't, it needs more heat. I did have to be careful not to get too much solder in the joint. I didn't want to plug up the hole.

This is the end result right after soldering. Looks pretty messy. I think I got the parts too warm with the map gas torch. If you get things too warm, the flux doesn't work as well and can burn and not do its job well. Flux's job is to keep the solder joint clean as the solder flows and bonds the metals together. In any case, I will have to pressure test the joint to be sure it seals up well.





And, here is one picture after I took a brass wire brush to it to clean things up a bit.





This wasn't a ton of work to do. But, with things being SO small, it is tedious. However, this is also the most difficult part of the build. So, if that is the worst of it, its really not that bad.

[Daox]

So many pictures. I need to host some in this thread for the previous post.

[Daox]

Making the jets

Now that we have our solenoid valve done, we need to make up some jets to regulate how much nitrous we squirt into our engine. These jets will screw into the ¼” npt to tube fitting that screws into the solenoid valve.





The jets will be made from #10-24 threaded brass set screws. Different sized holes create different jets, and different jets control how much horsepower we get from the nitrous system.





Once again, we will bust out the micro drill bit set that we used for the solenoid valve modification. This time we will be using smaller bits though. Reference the nitrous.info jet size chart to figure out what size jets you want. I am going to start with a 10hp shot which means I need a .4mm nitrous jet. This is an extremely small drill bit, and very easy to break. You must be very careful and very patient while working with these small bits.

Before drilling out the jets, I drilled and tapped the 1/4" npt tube fitting to fit the jets into. This fitting goes on the outlet of the solenoid valve and connects the nylon tube to it.







I screwed the set screws into the fitting which has been tapped for the set screw (#10-24), and then set the fitting up in my drill press vise. This will aid in holding the tiny little thing as best I can.





You can repeat this step for as many jets as you would like for different power outputs. I've had a heck of a time drilling the holes for anything under .6mm thus far. So, I'm still working on the .5 and .4mm jets. The drill bit kit I have comes with two sets of drills and I've snapped one set of .4 and .5mm drills already.


I recommend bagging the jets individually and labeling the bag after drilling them. If you don’t, you will never again know what size that jet is, and knowing your jet size is very important!

[Daox]

So, its been a little bit since I posted here. I haven't really worked on this project as we've been in super arctic freeze mode lately here. I need to get out to the garage to setup the drill press to get those last two jets made. I think I have a solution to my previous issue. But, for now I would rather keep my fingers non-frostbit. Hopefully, I'll warm up so I can continue on.

[7milesout]

Daox - I'm late in here. But I wanted to ad my $0.02. Might only be worth $0.01 or less. I'm a big engine / muscle car / turbodiesel fanatic. Naaaah, maybe I should just say I'm a motorhead, that's better. Most guys from my world would laugh and walk away, reading about what you're doing. The rest may laugh and then come in here and insult you. I don't want to do either.

First, I'm very interested in owning a Mirage, but I have a few other cars in the way right now. I own 6 cars at the moment. Most are just older cars I've had a long time. Like a 2000 Dodge Ram with a Cummins and 6 speed I've owned since 1999. And a 2010 Lexus LS 460L. But I really like the Mirage. I like cars that have a point. And to me the Mirage has a good point. It's point is to be INEXPENSIVE. I didn't say cheap, I said inexpensive.

When I get one, I will not do a darn thing to it but drive it and maintenance. That being said, props to you for this project. However, there is something you can do that will open your eyes, and lift the veil from what you're doing. An AFR gauge. One of the cars I own is a 1972 Plymouth Scamp. It came with a 318 V8, but a hundred years ago, the previous owner and I put a warmed over 360 V8 in it. A couple years ago, the previous owner gave me the car. He is in his mid 80's now, and for nearly 20 years, he's been having to give his wife full time care. So he had not been able to take care of the Scamp.

When he gave me the car, it needed … A LOT. The exhaust was completely rotted out. Amongst a hundred other things, I replaced the entire exhaust (from stock-ish manifolds to TTi headers, 2.5" pipes, X-pipe, Borla mufflers). During that time I debated on adding an AFR gauge. I went ahead and did it because that was the best time … when the AFR gauge could be added during the install of the new exhaust. It was about $300 extra dollars into the exhaust (IIRC). That's the AFR gauge plus labor.

It was, and remains, the absolute BEST THING I have done to the Scamp bar none. I can't imagine life without it. I have been able to tune the carburetor to the ultimate. People have ridden in my Scamp and can't believe how good it drives on the street, plus how much power it makes. It's all because of the AFR gauge. Without it, I would still be believing the engine was running rich. Before the AFR gauge I used to think I was flooding the engine and it would take me forever to get it running because I would stop pumping the gas pedal. Turns out, it was way too lean on cranking and at idle. After studying the AFR gauge, it seems as if I could fill every cylinder up with gas, and this engine would burn it and love it. Seems like there's no way to flood this engine. It LOVES gasoline. Anywho...

Sorry for the long post, but that was the background to say this. No way in hell, would I put a dry set-up on ANYTHING. It's only going to go lean. That's all it can do. Maybe modern FI systems can make up the difference somewhat, but I think it wouldn't make up that much of a difference. Here's my thinking...

Really the main point at which a lean burn is going to be a (big) problem is, at WOT. At WOT, I'm going to go out on a limb and guess that even a 78 hp 3 cylinder's injectors are going to either be at 100% duty cycle, or very close to it at WOT. Spray NX in there dry, it's going to go lean. And at WOT, that's where the AFRs are critical. If it even leans it out from somewhat rich to stoich, it may be safe, but there's not going to be much or any power gain doing that. Stoich is not the optimal AFR for WOT power. So … no way I'd spray anything dry. Before I installed this AFR gauge, I would have sprayed dry. Now, no way. I didn't get to read the whole thread, but I don't believe you're going down the dry path anyway. I'm just saying my opinion is to never do a dry system, and I 100% support your choice to go wet.

The other point I want to make is, if you forget about your $100 limit, and bite the bullet and install (or have installed) an AFR gauge, it is a simple affair (for you anyway) to spray the perfect ratio of NX and gasoline into your 3-tapper (it doesn't bang … it taps! ) And I call it a tapper with much respect. I love that 3 tapper!

For (probably) all gasoline engines, max power at WOT is achieved in the 12.2 AFR range. When adding a wet shot, it is the same ratio. 12.2 AFR at WOT. I dare say, that 3 tapper could add probably 50 hp and be perfectly fine. As long as the AFR is not lean, and achieves in the neighborhood of 12.2, I bet the Mirage engine is plenty strong for 125 hp.

I'm trying to achieve a 13.999999 second E.T. on the set-up the previous owner handed me (except an upgraded / new replacement exhaust and a lot of carburetor tuning). I'm at 14.178 seconds. And that was due to an early shift to 2nd gear. I think I'm there already, but want documentation (the drag strip slip) to confirm it. My Scamp is not an animal, just a really good running car. And it is on skinny tires, almost as little as the Mirage's. If I could launch hard, it would hit 13.99 easy. But I have to baby it on the start line. Once I achieve 13.999 on my current (original) set-up, on will go a new carb, new intake, get that dialed in. And then I'll be putting my money where my mouth is, and installing a wet system on my Scamp. Targeting a 150 hp shot at 12.2 AFR. It will be safe hp at those AFRs.

Anyway, that's my long winded reply to say, a) More power to ya! And, b) I strongly recommend an AFR gauge.

[phaux]

However, there is something you can do that will open your eyes, and lift the veil from what you're doing. An AFR gauge.
...
Really the main point at which a lean burn is going to be a (big) problem is, at WOT. At WOT, I'm going to go out on a limb and guess that even a 78 hp 3 cylinder's injectors are going to either be at 100% duty cycle, or very close to it at WOT. Spray NX in there dry, it's going to go lean. And at WOT, that's where the AFRs are critical. If it even leans it out from somewhat rich to stoich, it may be safe, but there's not going to be much or any power gain doing that. Stoich is not the optimal AFR for WOT power. So … no way I'd spray anything dry. Before I installed this AFR gauge, I would have sprayed dry. Now, no way. I didn't get to read the whole thread, but I don't believe you're going down the dry path anyway. I'm just saying my opinion is to never do a dry system, and I 100% support your choice to go wet.
...
For (probably) all gasoline engines, max power at WOT is achieved in the 12.2 AFR range. When adding a wet shot, it is the same ratio. 12.2 AFR at WOT. I dare say, that 3 tapper could add probably 50 hp and be perfectly fine. As long as the AFR is not lean, and achieves in the neighborhood of 12.2, I bet the Mirage engine is plenty strong for 125 hp.

Man, I thought I was the long-winded one! J/K

I've read up myself on Nitrous and it seems like the biggest issue with it is just watch the AFR and don't go lean. Is that really all there is to it? Besides stuffing the whole bottle down the intake

The Mirage has a Linear AF sensor in the exhaust manifold, I've been able to read a real time AFR from it, do you think it'd be safe to use that instead of having a separate gauge installed? Since it's already there and all...

Also speaking of duty cycle, from the other thread about the injectors Daox had them flowtested and got 230cc, which according to some math says the fuel system can handle 130hp at 100%. Using an app I found that at WOT the most it ever went was about ~60%, so it sounds like the Mirage isn't taking full advantage of the injectors (makes sense for a conservative OEM tune). It sounds like there's plenty of room for improvement here, but I wonder if you cram the Nitrous down the intake will the ECU actually add fuel to it or just go lean?

I hope I don't sound like I'm discrediting everything you just typed, I feel like there's multiple threads happening about different parts of adding performance to the Mirage (this thread, injector flowrate, ecu tuning/piggyback).

Edit: Best I could do for Injector usage was 13.8ms @ 6500RPM, which is like 75% duty cycle.

[Daox]

I agree with you 7milesout. AFR is very important when doing a mod like this. We don't want to be melting or detonating anything. As Phaux mentioned, the Mirage actually comes with a wideband O2 sensor. So, we can use it to monitor air fuel ratios and make sure that they don't get too out of hand.

For the time being, I am planning on starting with a small ~15hp dry shot. From there I will increase it and we'll see what I get up to. But, I will be cautious as I move ahead.

Thanks very much for the feedback! Its good to iron out these things before hand.

[7milesout]

Is that really all there is to it?

No. I'm no expert on the actual nitrous system. I consider myself to be an expert on normally aspirated Edelbrock carb tuning, but only when an AFR gauge is included. So, very limited. But I think outside of AFRs, I think a fair amount of modders get in trouble with leaking nitrous where it shouldn't go, and having ignition timing issues corresponding with nitrous usage.

The Mirage has a Linear AF sensor in the exhaust manifold, I've been able to read a real time AFR from it, do you think it'd be safe to use that instead of having a separate gauge installed?

Yes, and that's AWESOME if you can tap it and it read real-time.

Also speaking of duty cycle, from the other thread about the injectors Daox had them flowtested and got 230cc, which according to some math says the fuel system can handle 130hp at 100%. Using an app I found that at WOT the most it ever went was about ~60%, so it sounds like the Mirage isn't taking full advantage of the injectors (makes sense for a conservative OEM tune). It sounds like there's plenty of room for improvement here, but I wonder if you cram the Nitrous down the intake will the ECU actually add fuel to it or just go lean?

I admit, I could easily be wrong. It could be that Mitsubishi uses carryover injectors from some other application, and "tunes" the duty cycle to fit the needs of the Mirage 3 tapper. In that case, it stands to reason the injectors could make up a larger difference in AFR than my original thinking. But … I wouldn't know how anyone could modify the duty cycle. Meaning, first off, I wouldn't know how to modify the ECU. And secondly, even if one *could* modify the fuel mapping, without a crap load of time experimenting, I can't see how it could be modified accurately over a broad range of rpm. Carbs are easy … it's just geometry and hardware.

I hope I don't sound like I'm discrediting everything you just typed, I feel like there's multiple threads happening about different parts of adding performance to the Mirage (this thread, injector flowrate, ecu tuning/piggyback).

Discredit as needed. My experience is with carbs, normal aspiration, and AFRs. The NX experience will hopefully come by late summer, maybe over the next winter.

As Phaux mentioned, the Mirage actually comes with a wideband O2 sensor. So, we can use it to monitor air fuel ratios and make sure that they don't get too out of hand.

daox - When looking at the reading from said wideband O2 sensor, what exactly does it show? Does it read out an actual Air-to-Fuel number? In a range of 10:1 to 20:1? Something like that? Or is it other? I can read, I think it's fuel trim data, from all my cars (not the Scamp of course) via a BT OBD-II connector and an app on my phone. It displays a gauge that swings the gammot from more rich through stoich to more lean … back and forth. But I cannot distinguish which is lean and which is rich, nor can I distinguish by how much. So that's mostly useless.

[phaux]

No. I'm no expert on the actual nitrous system. I consider myself to be an expert on normally aspirated Edelbrock carb tuning, but only when an AFR gauge is included. So, very limited. But I think outside of AFRs, I think a fair amount of modders get in trouble with leaking nitrous where it shouldn't go, and having ignition timing issues corresponding with nitrous usage.

I figured there was other basic stuff like making sure your plumbing is good and probably having a controller of sort, but it sounds like really just watching your sensors and not starting with a 200hp shot is the key. I imagine things you need to watch would be the AFR, Timing, Fuel Trim, and Duty Cycle?

daox - When looking at the reading from said wideband O2 sensor, what exactly does it show? Does it read out an actual Air-to-Fuel number? In a range of 10:1 to 20:1?

Works exactly as you expect, spits out a decimal number (14.7, 13.5, etc). Thank god for standards.