running over 15 PSI? you need to know this.
 

with the proliferation of single turbos and AI more and more FD pilots are turning up the boost.

of course you have to have all the ducks in a row and if you do the rotary is easily capable of cranking out 500 rear wheel hp... reliably.

given the Combustion Chamber Pressure (CCP) envolved when you are making 2-5 hp per cubic inch there are a few things that can spell instant motor doom.

i would like to draw your attention to one of the most lethal and probable.

whether you run the stock computer or a Power FC both the fuel map and ignition map reference off RPM and Load.

Load is the key. Load is determined by manifold vacuum or boost by the MAP sensor. (MAP being Manifold Absolute Pressure).

as most know, the MAP sensor bolts to the firewall near the brake reservoir. it connects to the engine using a small hose. the MAP sensor(whether stock of the often used GM 3 BAR) has a really nice lip on it to ensure that, w the addition of a tie wrap, the hose will not be blown off the unit.

////edit: after writing the above and not recently seeing a stock MAP sensor i actually am not sure that the stock sensor has a good lip or not... the GM unit has a really good lip. someone please let us know re the stock sensor lip////////

the other end of the MAP hose is where motors are blown.

we generally connect the MAP sensor hose to the engine using one of the 3 tubes at the rear face of the UIM. that's the problem... they are tubes w no retaining lips.

consider what happens when you have upped the boost, your right foot is on the floor and the hose blows off the UIM tube...

the computer senses no boost while the turbo is still making, say 18 psi. the computer immediately moves fuel and timing to a vastly different spot on the map.

instead of 15 degrees advance the engine receives 30 degrees. instead of 75-85% injector duty cycle the engine receives 20 or 30% duty cycle.


we are at redline and all of a sudden we have double the ignition advance and a quarter of the fuel.

all because of one hose being blown off!

as you know, there are around 9 tubes that we either use or cap off on the UIM, none have retaining lips. most, if they fail, will not cause engine failure.

in the 21st century enviornment w the ability to run 27 psi on pump gas and AI methanol reliably all of us need to rethink our pressure connections. i have a 3 inch lab grade pressure gauge and an adj regulator and it is easy to pressure up the connections to 30 psi to check integrity. i was really surprised as i headed towards 30 psi to see so many tie wrapped connections fail.

i welcome anyone's comments as to how they have uprated the connections. the simplist, though not necessarily the best, would be to make sure there is no grease on the tube (wipe it w alcohol) and use multiple tie wraps.

please do spend a moment and recheck your MAP hose at the UIM.

howard coleman

Howard, is there a way to expand the opening of the tubes that come off of the UIM, thereby creating your own "lip"? If not, would some sort of epoxy be able to bond with the metal and rubber?

I suspect that a simple scuffing with sandpaper then laying a nice little bead of jb weld will work very well. You might want to sand down the JB Weld 'barb' a bit to make it easier to work with.

Another option would be to physically yank out the metal nipple from the manifold and tap in an NPT and use a barbed fitting.

FYI, assuming a .1" diameter hose, at 30 psi your 'blow off' force converts to .25lb. I suspect the jb weld method plus a zip tie is more than sufficient.

Dave

you can cover up the holes with jb weld. use a vacuum manifold and use fittings with a lip.

from the get-go I drilled/tapped a threaded port in the intake to use a different style connection. I didn't like the straight "no lip" connection point that was available from the factory.

setzep did it right.

since my manifold is off for winter i see some drilling and tapping in my future. you don't necessarily have to D&T all of the fittings as many of them won't cause engine failure.

your really do not want the MAP hose coming off.

guess how i know this.... it happened last Sep on the dyno. fortunately ace tuner Luke Stubbs got out of it w no motor damage. my hose had been tie wrapped.

i do not want to lose a motor to something simple to fix.

neither do you.

where' the drill?

howard

how can you get the the metal tubes out? pair of vise grips and twist like you can do with studs.

This is a really important point. On my 2nd gen my OMP air bleed hose blew off the back of the UIM. For some bizarre reason that distorted my GM 3 bar MAP sensor reading, which was hooked to a port on the front of the UIM. So I was boosting 21psi (race fuel), but my power FC only read 10psi! The timing was advanced too far by about 8 degrees, and the AFR leaned out to 15:1 at 21 psi!

The result was INSANE detonation. The porcelain separated from 3 of my 10 heat range spark plugs! I cracked a corner seal. But RA super seals survived and I am actually reusing them on my next motor as they look brand new almost and passed every FSM test.

One soluton: A small bit of Krazy Glue. Seriously. you will have to cut the hose off to remove it (MAP sensor hose isn't so bad to replace), but it will NEVER blow off. Good to 25+psi . For my actual MAP sensor, I actually use a 3/16" fuel line hose from autozone on my FC, I have it clamped with a small hose clamp on a nipple on the front of my 2nd gen UIM (smallest size clamp you can get there). The 3/16" hose will fit very very tightly on the GM 3 bar sensor (b/c of the lip on the nipple), nothing extra needed.

Here's a photo of the stock MAP sensor (Mazda calls is a Boost Sensor ). A little sandpaper and crazy glue sounds like a very good idea. Might be wise to make sure all the other vacuum lines on the UIM are secure also (especially the one connected to the fuel pressure regulator)... as mentioned above, any boost leak large enough to distort the vacuum that the MAP sensor measures could result in a pretty catastrophic failure.


https://www.rx7club.com/attachment.p...1&d=1235860917


-s-

scotty305 makes a real good point.

we need to add to the list of absolutely locked down connections the line from the UIM to the aftermarket fuel pressure regulator.

let's assume we are at max boost, say 25 psi, and that line pops off the UIM tube. fuel pressure goes from 65 psi to 40 psi. at 25 psi boost. ooops.

where's the drill and tap?

and thanks for the MAP sensor pic. that sure looks like an accident waiting to happen. CrazyGlue (along w tie wraps) sounds like it might work... anyone have an informed opinion?

hc

Another important thing here is running the RIGHT vacuum line. If it doesn't fit well to begin with, you'll always have problems.

I have always run 3.5mm Hose Techniques silicone vacuum line. It fits VERY tight on all the stock nipples, and requires a good deal of force to remove. Granted I don't run 20+ psi of boost (right now about 12-13 psi) but I've NEVER had a line pop off. I don't have any of them zip-tied either.

The FPR and MAP sensor lines are absolutely critical - this is definitely true. Same goes for any lines for a boost controller.

Few good things to make CERTAIN of. This goes for any FD, not just cars running a lot of boost.

- NEVER NEVER NEVER use plastic vacuum tees. I personally know of 2 people who lost motors due to these failing. The heat under the hood is just too much. They're fine on Grandma's Buick, not an FD. If you have to use a tee, use BRASS. For small size tees, you can find brass ones at the aquarium section of most pet stores for cheap, and they hold up and work/fit great. For larger size, like wastegate hoses (6mm), some parts stores might have them, but McMaster-Carr is a great source, they have really good ones for cheap.

- NEVER NEVER NEVER tee anything in to the MAP/FPR sensor signal. You want that signal as clean as you can make it. Also, tees, even using good brass tees, introduce a potential failure point.

- NEVER NEVER NEVER use cheap vacuum caps. The ones from the parts stores WILL FAIL. The heat bakes and cracks them, and that's it. McMaster-Carr sells good fitting vacuum caps in a variety of sizes. I've been using the EPDM rubber caps they make, good to 450 degrees, for years now with zero failures. They're dirt cheap for a big bag of them too.

Also, there's always a debate about using viton or silicone vacuum hose. I have worked with cheap silicone vacuum hose in the past, and it IS crap. GOOD QUALITY hose, like from Hose Techniques.com, will not let you down. I've been using their hose for 10 years now on turbo FC's and FD's and have had ZERO failures. I do use Viton line from the oil neck to my catch can - that is just way too much oil vapor for silicone line to tolerate, even good stuff. Viton stretches much less than silicone, and I'm not sure how "grippy" it is to prevent a hose from popping off. 3.5mm and 6mm hose from Hose Techniques does the trick - it's reasonably priced and will last the life of the car.

If you're running a single turbo with a very simple vacuum hose setup, I think it would be worth re-thinking the whole vacuum line setup. Remove and weld over the stock vacuum nipples that you don't need. If you run a GM 3-bar, drill/tap a 1/8" NPT hole in the UIM for a threaded brass hose barb that's 6mm size - the 3-bar sensor (if memory serves) takes a larger size vacuum line, and running some sort of adapter to go from small to large doesn't cut it. You have to stop and carefully consider each part of the system and how it could fail.

Dale

Excellent, excellent posts guys....this should be REQUIRED READING.
Now I am going to go out and at least check mine out and see what route to best take.

5/32" is about the closest to 3.5mm you are going to get in a nominal standard size from my experience, having bought hose that was listed in standard sizes. i think I am going to order 5/32" vacuum caps from McMaster-Carr, part # 6448K74 :

http://www.mcmaster.com/ctlg/DisplCt...54344087987953

and do not bother with autozone brand silicone hose ("3A Racing" or whatever). you can just feel how cheap it is compared to the real stuff from hose techniques. And I think zipties are worthless. I have done a lot of pressure testing of my whole air system and have found that they make no difference. You can't get enough torque on them--only worm gear clamps can do that, and you can't get ones that will fit over the small hoses from what I can see. The most important thing is tightly fitting hose.

I run my wastegate lines in stainless steel. It's the only thing I trust. You can get all the fittings and such or just use a worm clamp. it's a bit ghetto looking but it works, although be careful not to get pricked.

good stuff:) thanks for adding value to the thread.

hc

Years ago after a dyno session, the shop actually left the map sensor loose after the run. I got about 30 yards down the road before the car just crapped out on me. After that I taped it up until I got home and secured it propperly.

I used a small tapper and knocked the end with a hammer until it stretched abit of the end and created a little lip. Then used zip ties to secure. Granted its not as effective as some of the above but I've not had problems with it at 18psi :)

So what would be the best size caps to use? Would it be the 5/32" like arghx member stated?

Howard since you have the pressure gauge, try to use some of the grey hi temp gasket sealant and see how that holds up for us.

It may be possible to use a brake line flaring tool to make a lip on the end of the factory vac. line. The end of the tool is a cone and there is a die type device you may be able to clamp on the line and then use the cone to press a lip on it. I would recommend trying it on a junk manifold or something else just in case it does not work. You can find picture of the tool online with google. It would be interesting to see if it does work. I havent heard of the lines blowing off before, I have done many 22psi nissans and those are straight non flared vac fittings. But if it does work I would do it just for the extra peace of mind. Let me know if some one does try it and it does work.

I will let you know when I get the 5/32" caps in. Like I said I have used 5/32" hose before and it fit very tightly.

I used to use a little red loctite on the nipple and then push the silicone line over it. It would make the line VERY secure on the nipple. I would have to cut the silicone to remove it most times. Just make sure it cures before doing max boost type runs. Superglue is similar but dries too fast for me. Going to straight AN fittings would be best in most cases (especially at the wastegate) but how would you attach an AN to your map sensor?

In my case I lost the engine when the banjo bolt on the wastegate came loose and fell off at full throttle. The hose stayed on the nipple though :) Funny part about that is I had ordered the AN fittings to re-do the wastegate lines because I was paranoid. They got to my house two days after my engine blew up... talk about bad irony.

Great post. On a lot of FDs, the hose leading to the MAP sensor is baked.

I had the hose blow off of the map sensor side on my old FD. This happened while it was under load, too. Scared the hell out of me and I was sure that I hurt something. I had a J&S Safeguard in that car (set to pull a max of 10 deg to avoid negative timing split), and it went crazy with pulling timing when the hose came off.

I sandpaper'd the nipples and used Hose Techniques hoses. Never had a problem after that. This was with 12-13 psi of boost. I might not have been so lucky with more boost.

Sonny

this is great info as I am currently switching my hoses around. Today i was installing my new fuel system, and the comment about NEVER t-ing into the MAP or FPR lines got my attention as I was just about to do it.

The two hose fittings at the base of the LIM that are burried by the fuel system are impossible to get to without removing the fuel system. For this reason, I was going to remove rear fitting that goes to the stock FPR, as my new FPR will be in the front. So I was going to just cap that fitting off (no barb uh-oh).

The front fitting currently goes to my boost controller, and is right where my FPR and boost controller will be. So I was going to T into this for both. If this is a no-no, and I want to avoid using the rear fitting which is completely buried, what is the optimal spot to locate the boost controller source (I'll leave the new FPR hooked to the front fitting that is buried). Is it safe to T into the hose that goes straight from the UIM to the Blow-off valve? Its in the same area and would be easy to get to.

And I capped the rear nipple off today before i read this thread. I rubbed high temp orange RTV on the fitting, put a pretty thick cap on it, two zip-ties, and then a tine squeeze hose clamp (like a radiator clamp that works with pliers). I don't know where to find a worm clamp that small anywhere. I won't be running over 24psi, i have not finished the fuel install yet, so i can go back and redo it if needed, not sure how i could do any better though.

This thread makes me think that a MAF (mass air flow) based system would have been better. I never really got a good argument as to why MAP (manifold pressure) based systems were used in place of a MAF type... especially on a turbochaged car... Anyway, good points on how to ensure system integrity... I echo the statements regarding securing the MAP line and the FPR line.

Some more modern engines...like the .. cough LS1 cough.. Run both a MAF or HFM and MAP sensor. All in good fun :) A lot of the V8 guys run speed density off of just the MAP sensor but it isn't as precise. IT is strange that an engine as reliant on tuning (they all are, but the rotary moreso) as the 13brew, would run only off the MAP sensor.

1. 2nd gens had both an airflow sensor and a MAP sensor from the factory (and EFI 1st gens had just an airflow meter). It has nothing to do with how modern or un-modern the car is.

2. running just a MAP sensor in speed density configuration cut down the cost on a car that was already really expensive and complex.

3. Mazda does not design their cars with hardcore aftermarket tuners in mind, nor should they have to.

I understand your points but, particularly regarding 2 & 3, when there is such a narrow margin of safety, especially in warm regions (or when the car is being pushed on occasion like an FD should), on operating a high output "rotary experiment", why would you--as an OEM--compromise? No wonder the FD has such a bad rap.

Who says anything about hardcore tuners, that was not my implication. Even in stock trim, one can argue that the car is thermally mismanaged! I know that competing factors like performance, cost, weight, etc. all had something to do with the final product... but that doesn't mean anything to us, the consumers of the product, when we all know that the car is truly a thermal mess in stock trim. Heat is essentially a way to accelerate aging processes. Given that the rotary is a very good heat producing engine (and not a good heat rejecting one!), it doesn't do us any good when you put a turbocharger on it. I would like to think Mazda had their best guys on this project, but there's no denying that there were too many competing factors that led them to the car we all know and love, and sometimes are frustrated with. Any high performance motor is all about keeping heat in check... keep it cool as they say.

I personally feel the motor has enormous potential, it just needs the TLC the Mazda engineers couldn't afford to give us. Even if you're at 300 whp, you could probably use water injection or really high octane fuel if you're pushing it on a 100F day. Never did the calcs, but from my anecdotal type recollections that is about right. But what do I know?

:icon16:

A couple years ago you would rarely if ever find me defending Mazda or OEMs in general. But after being exposed to design principles in an engineering curriculum and doing some minor work with GM designers on a project, I have a new found understanding and respect for the difficult choices that factory engineers make.

I think Mazda made a lot of mistakes too, especially in terms of the reliability of a 100% factory vehicle. That's the main type of reliability you have to judge here. Even though it would be nice to have added safety margin and more overall easy modability, you can't blame a cost-conscious OEM because its car doesn't do what you want it to do when it is operated far outside its design parameters. It's not like Mazda was going to change its manufacturing process. It would have to use different barbs/nipples on the manifold and MAP sensor, adding cost to the car while providing zero benefit in reliability to anyone who operates the vehicle the way it was intended to be operated.

How many 100% factory Rx-7's, even poorly maintained ones, suffer engine damage from a MAP sensor hose blowing off (without the owner/mechanic screwing something up)? It just doesn't happen. Even though plenty of other different types of failures occur to these vehicles when stock, that's about as idiot-proof of a MAP sensor design as it needs to be. How often do you even hear of stock cars with MAP sensor related problems at all?

How many factory cars detonate because the stock ECU is speed density (no MAF sensor) and isn't mapped for breathing mods? That's what the fuel cut is for, to save the motor in case of some kind of freak accident on a stock vehicle. The car was expensive enough as it is, adding those features (better hose connections or a MAF sensor) would neither increase sales nor reduce warranty claims. We can go open season all day on the heat problems of a stock car, but the speed density issue and the MAP sensor plumbing are IMO beyond reproach from the perspective of a factory engineer.

In your defense however: Mazda went with a MAF sensor on the Rx-8, streamlined the emissions systems and eliminated many of the associated vacuum hoses, eliminated most of the idle problems with the drive-by-wire system, and drastically improved the life of the catalytic converter as well. So they recognized their mistakes, except for the Rx-8 OMP oil injector issue which was then corrected in the 09 models.

can anyone please help me out regarding my post on the first page? I would really appreciate it, i just don't want to have to take it apart again after i button it all up. thanks :)

Again, agreed on all counts; you're preaching to the choir on this one :)

Back to the OP topic: make sure your air & fuel lines are tight and right. Use good gas and make sure to at least run a good IC or WI (or both) if you're running 15+ PSI boost, stock turbos (especially!) or "effecient" single. Don't run crazy timing. Keep your fueling rich. Bingo. Reliable motor :)

McMaster Carr rules....

I think it was mentioned before, but without comment: use a vacuum manifold.

Comments on that by someone that has used them?

It appears to be a cheap solution to the problem of sourcing vacuum.

Here's one example: http://www.maperformance.com/vibrant...ion-block.html

I have to say that some of the measures being taken are getting crazy.

Yes, it is a critical connection, I value that. There are many critical connections on the car.

Notice that a .1" diameter hose at 30psi only generates .25lb of force to pop itself loose. That's four ounces, for 30psi of boost.

With stock hoses the rubber tends to harden, and the thermal expansion/contraction causes the hardened hose to fit loosely. Obviously this is unacceptable, which is why everyone should have a better material already in place.

A good silicone or viton hose on a clean plain metal nipple requires several pounds of force to pull off. There is ample holding power available. Burnishing the finish of the nipple or using an adhesive will make this even more secure. In my mind, it's as simple as noting how secure the grip is when I'm working in that area.

If you want mechanical retention, flare the end of the nipple or tap and install a barbed hose fitting. A zip-tie with a barb is extremely robust.

I have no problem with people going to extra lengths, belt-and-suspenders engineering, etc. But to say that a proper silicone hose is insufficient is going a bit far in my opinion.

Dave

Commenting on my own post here, as I waited too long to edit it:

One thought that comes to mind regarding this is that it would simplify removing, installing the UIM, and also make it simple to see if all the vacuum hoses are connected properly.

When you get much over about 17-18psi, all your silicone or basic rubber hoses expand a bit. Keep that in mind.

I got my 5/32" vacuum caps from McMaster-Carr (part # 6448K74 ) in today:

https://www.rx7club.com/attachment.p...1&d=1236195765

You can see they're a little bit longer than the autozone ones you typicall buy. I'm going to cut a little bit off one end, it shouldn't be a big deal.

These are actually a high temp rubber, rated to 450 F according to the McMaster-Carr site. The silicone ones are rated at a higher temperature but they only come in oddball sizes. These fit nice and snug on the OEM vacuum nipples. I will secure them further with a bit of Krazy Glue.

All good points Dave. The force as a result of pressure acting over the hose area is very low as you mentioned above. As far as thermal expansion goes due to engine bay heat, which would surely outweigh manifold air temperatures...... well let’s just say it’s less than a .001" added to the original hose diameter. (If considering silicone hose and average engine bay working temperatures)

It seems the factors against us are not that great. However over time and heat cycling material performance begins to degrade, as we all know. Of course some materials like silicone handle these factors much better than others.

All and all I'd have to say if the risk is great (like when a hose pops of the MAP sensor) then why not take the time to make sure the certainty of that risk is reduced.

Great catch on securing these two critical lines.

Years ago I hooked my boost gauge MAP sensor into a tee off the ECU MAP sensor line. I used a forged brass tee, Hi-temp silicone sealant etc. and was thinking I wanted to measure boost exactly where the ECU measures it and that a failure in boost line would thus indicate on my gauge. I did not think about the consequence of me added an additional failure mode.

In preparation for the trip down to Deals Gap I performed a lot of maintenance this past weekend. This included cleaning up my engine bay, using this thread as inspiration :)

I deep-sixed three plastic vacuum tees (y's really) and gave all the important connections their own dedicated vacuum/boost source. This included the FPR, MAP sensor, and water injection boost solenoid, which were previously T'ed.

Now there is one sole vacuum Y, for the boost gauge and boost controller. It's a robust stainless steel piece. There is one plastic vacuum fitting, to connect the 6mm BOV nipple to 4mm hose to connect to the UIM. I plan to switch that out when I have the proper fitting.

I'm using Sard vacuum hoses, and I'm extremely happy with them. They're over 2 years old and are still in extremely good condition, they also firmly attach to all the nipples on the UIM.

As stated I performed a lot of maintenance at once (new plugs, new NGK race wires, Amsoil tranny/diff fluid, fresh idemitsu engine oil) so hard to say if this directly affected driveability at all. I will say that the car definitely seems smoother, a result of all of the aforementioned I think :)

Edit: Also installed the Rotary Extreme hood struts, which I freaking LOVE :D

https://www.rx7club.com/attachment.p...3&d=1238984847

What a good read.... Howard Coleman is the man!

Here is where I am confused, does the FD have a TPS (Throttle Position Sensor)?

Yup. backside of the TB.

Just want to re-iterate from first hand experience... The FPR vac line is NOT barbed
http://sphotos.ak.fbcdn.net/hphotos-..._2100142_n.jpg

Lost boost reference @ ~15psi. Currenty awaiting a compression check.

Just jumping on! Bump!

I just ordered some 4mm hose clamps from ebay. Wanted to try them out on some connections that I have. I will let you guys know if they are a good fit.

Same exact thing happened to me. My hose blew off doing a quater mile pass and I ended up loosing 3 corner seals. However, my RA Super Seals lasted.

I now use clamps on all hoses and super glue.

bumping a good read, I think I may have lost my motor due to a vacuum hose slipping off.

I have used both, and the Viton is definitely "grippier" than the silcone.

:-) neil

Bump for good info. I will make sure to keep this in mind when I do my single swap.

i agree, the FD in stock form, with all of the updates, is actually quite reliable. plenty of "normal" daily drivers around here with 100-130k on the original engines, there is even one locally that went 200k on the factory motor.

i do recall back in the day, the techs would occasionally bump the MAP hose off doing an oil change, but we never lost an engine from that. why? the car starts, but doesnt run well enough to drive!

the FD main cat is like the strongest one ever built, its a tank, you can blow the engine @100,000miles and drive with 1 rotor for another 50,000miles (like BATMAN), and its not dead. those things are TOUGH.

with the Rx8, the cat got tested during the (42006? 40206?) recall, the one that we checked the engines in late 06? the procedure was to test the engine, and then the cat. our dealership had about 150 rx8's in service, and out of those we replaced 0 engines (better than the FD) but many main cats (20 something? they weren't ordered thru the normal parts channel, so its harder to tell), about 1 in 5 rx8's had a bad (plugged) main cat.

i think looking back both cars are still probably better than the V6 mazda 6... we did more engines, transmissions and cats on those things, than either the FD or the Rx8, in the warranty period

anyways, always interesting to correlate perceptions with some kind of metrics

I have accidentally gotten Krazy Glue on silicone hose and had the hose crack open very easily around the area with the hard cured glue on it.

I use a flexible adhesive like weatherstripping glue and a zip tie and so far no failures.

I agree removing the stock nipple and drilling/tapping for barbed nipple is a really good idea.

I always thought it would be really cool to use something like a Swagelok SS-6M0-1-2 fitting swagged onto the nipples and braided AN vacuum lines.

I have been interested in these 3.5bar map sensors Zeitronix sells, but not sure if it's way overkill, that plus I really don't want my engine to be the guinea pig to test one. Anyone happen to have PFC scale/offset numbers for a 3.5bar handy? http://www.zeitronix.com/installatio...stallation.htm

Can use either screw on the the hose barb fitting or use a 1/8NPT bung and mount it directly on the intake.

http://www.j-specperf.ch/images/zeit...ure_sensor.jpg

Howard, you talked about drilling and tapping the UIM for the MAP, why not go another step further and tap for the FPR? And for the hell of it, couldn't a dedicated tap for the wastegate(s) be added as well? Seems to me, It wouldn't be a bad idea to at least do the MAP & FPR.


I have an idea Howard. For the squemish, couldn't you drill and tap the block off plate for the, AWS solenoid and Pipe? That wouldn't be all that difficult to remove and tap. Wouldn't be a bad location for the source either, I wouldn't think.