I have one.
It ran good for one set of races and then started acting up on me.

Here is the problem.
When I engage the system and floor it, the n2o turns on and it works briefly and the fuel pressure goes sky high(100psi) and the engine stalls until I let off the gas and it regains.

Now for the background.
The dry system works by letting a bit of nitrous pressure bleed off and ups the fuel pressure to compensate for the added air coming into the engine.
For a 75hp shot, the fuel pressure is supposed to go up to 70psi, but the fuel pressure keeps going up and floods or stalls the engine.

What I have done to correct the problem is..
new fuel pump
the return fuel line is not plugged
teflon'd all the n2o fitting in the box
narrowed the gap to .030 on all plugs
checked the n2o nozzle
checked the spark(hooked a timing light inside the car to make sure I was getting spark)
I have put in bigger injectors(neither the 19's or 24's work)
disconnected the o2 sensor
tried a different fuel pressure regulator

A couple other things I am going to try...
I got a msd digital 6 igniton and a blaster ss coil to put in(ignition related maybe??)
inside the n2o box, there is a brass block and within that there is a brass orfice, could the orfice be worn and opened up more ????)

Any help would be greatly appreciated!!!

I am at my wits end with dry nitrous system.

 

Get a wet system…
(never saw the point in running dry)

How does the ZEX setup control how big the hit/fuel enrichment is? Usually most dry systems have a jet in the line going to the FPR to limit how much pressure it sees and control the amount of fuel pressure increase, in that case you’d put a smaller jet in.

Consider yourself lucky, usually when you get many GM injectors into the 85-100psi range they start locking up and you start having rather unpleasant consequences, going rich and just not running right is a much better way to go then going boom!

 

It uses the bottle pressure(usually 700-1100psi) and lets some pressure bleed down through a tiny orfice and push on the fuel pressure regulator. There is a jet in line, like you said.

Does anyone think it is possible that the orfice has worn bigger and is letting more pressure up the fuel pressure??

 

Anyone have a zex dry nitrous system?

 

Here is the problem with dry setups that I can see. The fuel injectors are pulsed at a certain rate. The N20 is one the entire time so it never seems like a good mix of N20 and fuel. Basically, it seems like there are a lot of lean time. I don't like dry systems because of this.

I have been wanting to try a dry setup for a while now using this setup:
42#/hr saturated injectors
under intake direct port nozzles for N20
two N20 solenoids, one for system on/off, one for nitrous control (PWM)

Using the N20 with PWM would cure the problem if the ECM controlled the N20 solenoid and the fuel injectors. This would give a simultaneous spray and a good mixture. BUT, there is another problem and that is the rate which a N20 solenoid can be PWM controlled. Most can't be pulsed faster than 20 HZ.

That means that the N20 solenoid can only be turned on every 50ms. An engine running at 3000 RPM does one complete engine cycle (2 crank revs) at a rate of (120/3000) = 40 ms. If the N20 jets were big (only 100-200HP) range and the driver only wanted 50HP at 3000 RPM then the N20 should only be on for 1/4 of the time for a 200HP jet. That is, 1/4 time of 50 ms is (50/4) = 12.5ms. So, to get 50HP with 200HP jets then the N20 is on for 12.5ms. But the engine does a complete power stroke in 40ms at 3000 RPM so most of the nitrous (a true 200 HP shot) is given to mainly 40/12.5= 3.2 or 3 cylinders and a little bit to the others.

So, if jetted for 200HP and 50HP is wanted at 3000 RPM, then the engine really sees a 200HP shot go to only some of the cylinders. Not a 50HP shot to all of the cylinders.

Make sense?

 

Well, the idea is that you’re supposed to raise the pressure enough to add as much fuel as you would use to mix with the amount of N2O you’re spraying, the fact that the injector meters fuel by opening and closing shouldn’t limit that. The issue really becomes that since you might spray from 3000-7000rpm across 3 gears with different loads on the engine, that injector DC will change across the whole pass. Lets say that your injectors are open enough that you are making 400hp peak at 40psi NA, you’d think that for a 100hp shot boosting the fuel enough to raise that to 500hp worth (incidentally 62.5psig) would work, but if again, that power band is 3000-7000rpm, you’re probably seeing that peak at 6500rpm, and the injectors are delivering only enough fuel for say 150hp at 3000rpm, that same pressure increase is only enough fuel for 37.5hp, but you’re still delivering 100hp worth of N2O. Can you say BOOM???

What ends up happening real world with dry kits is that the pressure is basically spiked to where you’re getting 2x or more as much fuel as you really need for most of the power band, hurting power, washing down your bores, diluting your oil…

That’s when you’re getting N2O at all. Most dry kits mount the nozzle a few inches in front of the TB, so when triggered, the injectors dump fuel while the air already in the manifold is burned off (basically going rich by about 2x the amount of fuel needed to mix with the N2O but without the N2O, in other words, that same 400hp engine with a 100hp hit will get >600hp worth of fuel for a split second while it’s only getting enough O2 to the cylinders to make 400hp, something has to happen to the extra fuel, it sure doesn’t help power or the engine any) before even seeing the N2O hit at all, and then after seeing it probably not seeing it evenly.

People make all sorts of arguments against wet systems, but very few are valid. Basically the nozzles are designed so that the high pressure N2O “shears” the fuel stream coming out of the nozzle atomizing it and mixing it, so the mixture that each cylinder sees is the proper ratio no matter what happens. Pudding and other BS, no matter what the manifold design is, is a non issue if the “rules” are followed when spraying (only at above 3000rpm and when the throttle is WOT), you will not get puddling in anything under those conditions.

Not really what happens… again, pulse width and timing vs the intake valve being open isn’t really an issue, the idea is that you’re supposed to end up with a roughly homogenized mixture in the intake and when the valve opens that mixture is what is sucked in, not what is being sprayed at the time. AAMOF, sequential systems specifically do not spray when the matching valve is open, only when it’s closed.

FWIW, I’ve played with a JNMM on my WS6. The way it does it’s modulating is to pulse at a duty cycle relative to the full delivery rpm. In other words, if you have a turn on rpm set at 3000rpm, full on rpm set at 6000rpm, when it turns on at 3000rpm the DC will be 50% of the full hit (since 3000is 50% of 6000), 75% at 4500rpm…

What I usually did was put the biggest jets I dared run in the nozzle (a lot of the time it was the 250hp pills), turn the “full on rpm” all the way to the right (the max is something way higher then I’d ever see, want to say something like 14000rpm), set the turn on rpm and go. As I wanted more of a hit I’d turn the full on rpm over to the left and I’d get a larger percentage of the 250hp the closer that full on rpm got to an rpm that I’d actually see.

FWIW, I’ve never actually measured how fast the thing pulsed the solenoids, but it did work quite well (if you can get around the rat’s nest of wire), even with the giant NX solinoids. On initial testing with 75hp pills the spray was ramped in so gently that I couldn’t feel the difference at all, but the car ran faster. I ended up rigging a light into the solenoid circuit that would flash with the N2O solenoid to tell me that it was working since I couldn’t feel it for sure unless I was using a big hit, you couldn’t see it watching the car either, it was just like the car suddenly found a bunch of HP NA somehow…

 

Good post. It got me thinking a little bit differently on this topic.

All true. As an example, my rig has BPWs of 3ms at 3000 RPM and BPWs of 8ms at 5000 RPM and higher. If the fuel pressure was raised to a higher constant pressure for the 100HP at max. BPW then it would be lean below 5000 RPM. The BPW is more varied according to VE or MAF tables and RPM more than load though.

It sounds like the problem that the person posting sees.

Wet systems are definitely the way to go. I believe that the N2O shears the fuel and gives a good mix.

I want to try a dry setup for two reasons. One, I want to do it under a HSR manifold and have it completely hidden (drilling the back of the block). With the single N2O lines it will be easier to route. Two, because I am cheap and don't want to buy all the fuel stuff when I don't need it. It will go on a 350FWHP engine with 730 ECM/$8D or $58 code, Walbro pump, 42#/hr injectors. No extra fuel lines and stuff and I don't want to have to control a fuel solenoid too (that will wear out). I have been playing with the unused PWM drivers on the 730 ECM and it seems that they can be slowed down enough to run the N20 solenoid. I FET driver is still needed for the solenoid.
Using the ECM saves the cost of an aftermarket PWM solenoid controller. I can control the solenoid based on RPM and TPS.


You are right about the sequential injector firing on closed valves. As for the batch fire with the nozzle in front of the TB I see a little differently. Suppose the engine is 3000 RPM (40ms of time) with a street cam so that the intake valve open time is about 30%. To simplify things, suppose the engine is at the start of a engine cycle (2 crank revs) then the intake valves open 5ms apart from each other. If the N20 was was setup for a 200HP shot and 50HP was desired then the solenoid should be on for 1/4 of the engine cycle or (40/4) = 10ms. So, the first valve starts to draw in some of the 200HP shot being sprayed, then 5ms later the second valve starts drawing and so on. By the time the 4th valve draws it is 4*5 = 20ms later and the N2O was off 10ms ago. So there isn't that much N20 left hanging around for the remaining cylinders.

I want to go direct port N2O though so I think it changes a bit because the cylinders can't rob from each other as much as the "before TB nozzle" setup. With a direct port N2O each of the jets would be 25HP jets (total of 8*25HP) = 200HP. I would think that each cylinder keeps more of a (200HP/8) = 25HP shot than a spread out 200HP shot. Not sure, does that make sense?

The strange thing is that if I spray the 25HP jet for 10ms at 3000 RPM (40ms of time) with a 30% valve open time (.3*40ms = 12ms) then I am spraying for 10ms/12ms = 83% of the time. 83% of 25HP is about 21HP. HHHHmmm, this is starting to make sense. Maybe I should be spraying for a percentage of valve open time. That is, spray for 1/4 the time for the 12ms valve open time. Spray for (12ms/4) = 3ms when using a 25HP horsepower jet per cylinder and I want (25HP/4) = 6.25HP per cylinder at 3000 RPM.

At this point I am kind of confused.

That is exactly what I want with the PWM controlled N2O........smooth controllable power. Hard launchs break stuff. With a setup like that the time is made up down the track instead of at the starting line.

You were running a 250HP jet with a before TB nozzle that was a wet system? Did you ever notice plugs not reading consistently from some cylinder getting more of a shot than others? I have been thinking of using a 200HP shot. I am undecided if I want to jet it at the N2O solenoid or the runners. I planned on putting the nozzles upstream from the injectors to try and get a good mix.

This may be getting off topic, but in some way I think all this info will help the person posting with the Zex dry system. He will probably sell it.

 

VE and MAF tables are used to calculate load… I’m not sure what distinction you’re trying to make.

It’s a “feature” of running a dry shot…


Go back to the first thing I said in this thread, that about covers what you need to do to fix it. To make it better… you could try a smaller jet, but if you’ve got the correct one according to ZEX then you’re asking for trouble.

To answer the other question… no, they do not wear… the only way that I could see possibly getting more flow through one would be if it wasn’t properly seated, but I’d be shocked if you pulled that off AND somehow kept it from leaking out the line altogether also.

I think that both your reasons just amount to “you want to **** around with it.”

FWIW, at the speed that the JNMM pulses things I found that the NX relays work quite well (I’m pretty sure that the NX solenoids that I used were too big for the JNMM to trigger directly, I believe the circuit was designed for something more like the NOS solenoids, but I never actually tried it). Makes things a little easier then using a FET

FWIW2, originally I tried using some bosch style relays (can you say “Mark’s a cheap bastard”) and got screwed. When pulsed the bosch relays weld their contacts together. First street race (not really but I’ll call it that) that I got into I killed the other car and then my fuel wouldn’t shut off and at the time I couldn’t figure out why I was running super rich to the point that the car wouldn’t run, ended up getting the car towed home.

Yea, but it doesn’t really work that way… that’s the reason why you’ve got a plenum, what was sprayed diffuses into the plenum and you get a nice, even hit… you’d think it would be an issue but you wouldn’t believe how smooth it ends up being in real life. Like I said, no one could tell on the small shots with the big jets, inside or outside the car.

In theory, yes, in practice from experience, no.

I’m not sure that I understand why you’re confused.

My .02… you’re making this way too complicated in a way that you’re not going to gain a damned thing. To make this work right you’d have to not only get the circuit controlling the N2O ‘noid like you’re describing here, but really you’ll have to add fuel by doing it in the code through injector pulswidth, something like a second table that it only uses when the N2O is running, and then when you get that right:

I’ll bet that you won’t be able to document any difference in performance between that setup and what I was running.

FWIW, a hard launch breaks stuff, but you can break stuff NA or at part way down the track. I had one of these easy launches that I made it the whole way down the track, ran a mid 11@almost 120mph and in the process toasted the clutch and broke an axle _during the pass_ (the axle tried to come out when I tried to turn into the turn around, caught the caliper, bound and I spun… then when I got it off the track I found that I had to limp it up the return anyway because the clutch was slipping too bad to really move the car below about 3K rpm)

Yep, you can see the nozzle here:


It’s actually drilled angled in the MAF so the plume is centered on the TB airfoil and I didn’t really go after hiding it, but by the time I was got it looking neat the lines going to the nozzle and nozzle were in some split looming run under the stock fuel lines and the neatest place to put the solenoids was hidden partially behind the headlight with all the lines in the fender (I hate the look of braided line and an fittings if you only have a few bits of it under the hood). You couldn’t actually see any of it without digging around and actually moving stuff, but it still only took 30 seconds to pull the lines and change jets if you wanted to.

I never had any evidence of distribution or mixture problems with the setup, but I suppose if I did I could have just re aimed the nozzle and the JNMM actually has an adjustment for the fuel/N2O ratio so you can adjust the relative PWM slopes….

FWIW, I moved this whole thing over just like it was here to the TPI car and everything worked exactly the same way, I just didn’t take any pictures. Well, the nozzle was mounted on the passenger side a little further forward (probably 10” or so in front of the TB, right after the bend in the firbird ducting), and the ‘noids and lines were all mounted right above the FPR and run straight down the side of the engine it looked really cool and used MUCH less line then the LT1 setup

 

Me either. I guess I was thinking that you were thinking about the MAP sensor when you said load. Anyway, a knothead post on my part.


Yeah, either it is rich or lean using the stock ECM and a simple fuel pressure booster. There is not way around it.

I meant wearing out in terms of it being controlled via PWM. I would think the seal would wear out quicker. For the N20 I plan to run two in series due a possible PWM solenoid failure of a FET failure causing it full on.



Yeah, that is probably the main reason. After working on a $58 3-bar MAP code routine a few weekends ago I got a decent handle on the fueling equation. I realized it isn't that bad to add another table for the N2O to get the BPWs right. I need to test out and measure the ECM PWM freqs but they looked to be sub 50Hz when flashing an LED. I have been wanting to try this fore a while, but the cost of PWM controller steered me away and I don't want the big N2O hit all at once.

Most of what I have read is that all of the PWM controllers out there pulse them between 20Hz to 50Hz. Some are adjustable for the size of the solenoid. Smaller solenoid get do faster PWM so the resolution is better and theoretically smoother power application.

Yeah, no way will a mechanical relay handle voltage generated by the current draw of the relay. Did you have a diode across the relay? Every time you open the relay it wants to weld itself due to the large current draw of the solenoid (just like an old points setup w/o the capacitor in place).
I am worried about the FETs croaking and leaving the N20 full on. To solve the problem, I am going to have one PWM solenoid and one on/off solenoid in series. The ECM plenty of outputs to control both. I might even feed the PWM FET signal back into one of the ECM A/D channels to monitor if the FET goes t*** up. It all depends on how ambitious I get. It is a winter project in the making....so I am checking the mechanical feasability of it now.

Hey, at least you won. Didn't you have the power connected to a switch in the car to disable both solenoids?? Or did you have to pop-out the relay?


I hear you about the plenum. I just can't see it just yet. I think the whole engine has a filtering effect even if more N20 goes to a few of the cylinders. I was more worried about hole to hole strain, but from what you described it does seem to be the case.

I was confused because using the 3000 RPM (40ms) time means I want the N20 on for 10ms. Using the valve open time (12ms) showed that I would want the N20 on for 3ms. They don't match.......therefore the confusion.
I looked at it from a fuel required for HP point a view and think I have a handle on it now.


Yeah, I tend to get things theoretically correct before building anything. I hate getting stuff and throwing it in the corner because it didn't work out. I guess I go a little over board sometimes. Yes, a new table will be added for the extra fuel needed. One ECM digital input will be used for the N20 arm, one one A/D input for the PWM FET status, one PWM output for driving a FET, one discrete output for driving a FET.
In the end, I need to figure out the equations for both the N20 on time and BPW extra on time because I hate messing with stuff to get it to work right when I can predict and adjust ahead of time. I don't like saying that it works by playing with it along with a little magic.

It will most likely be the same. I have to have it all completely hidden though. I like the idea of the simplicity of one nozzle though (and cost too). I want the hidden system with the PWM so that it looks & feels like NA. From what you wrote, that is exactly what I will get if I don't screw it up somehow.
That setup you put on looks good. I never tried one of those air foils. They always seemed to block the intake breathe hole and I figured it would mess up the IAC (the LT1 intake has it's own IAC runners). I think turbocity had them for $15 or something. You think it is worth it?

 

Got the problem solved!!!!

I feel stupid but here it is..

I bought the kit from a Mustang owner

I called zex tech line and they said it will work on my tpi no problem


WELL!!, This is not the case. After a very long time I called the zex line again. I talked to a person there and I was WAY over his head with the operation of the system than him, so he sent me over to Patrick(r&d, basically made the zex system). He told me that the mustang jetting was way too rich for my tpi. He told me to use the LT1 jetting charts. I did not have a lean enough jet for the LT1 system(needs a 82 fuel jet) so I used my 52 jet.

It worked much better!

He said what was happening was the injectors would only work till around 80psi then static lock open, therefore closing them and then the fuel pressure would go sky high b/c it backed up the system.

SO, I need a leaner fuel jet and I should be better(I hope!)

Thanks to Patrick at ZEX!!! He knows his sh*t!

I will post times in July when I go to my 1st 1/8th mile race