'99 Mustang build

[BRN2RUN]

I'll update all the things that i've done to the car, previously (including the porting/ polishing work on the intakes, and the thread of the ongoing head work that i'm doing).

Today, for a fun project, i'd started work on a DIY ram air duct that goes into the passenger's side foglight area. I'd done one on my T-Bird SC, and it looked like it had raised the PSI a small bit at higher speeds). I've hooked up the front air collector duct, and bolted it to the bumper/ foglight area, with some 3" tubing. The front part of the duct worked out great--i'd gone to the hardware store and bought something called Stealth Flow for around $20, which is meant to drain eavestrough ducts, but the cool thing about it is that the mouth/ funnel of the system, when cut to a certain length, fits almost exactly in the front foglight area, and then you just cut a little bit around it to make it blend in a little better. The contours of the foglight opening seem to look like they'd be beneficial for directing air in, too. For the piping, I used Mole Pipe; also for eavestrough drains, because it otherwise maintains a straight shape, but will bend to the exact shape that you want to position it in.

I was mulling over what to use for a filter box area, and don't laugh--a 2 litre plastic pop bottle seems like it's going to work out pretty well. I needed something cheap and easy to cut, and something preferrably see through, so that I can see when the filter needs to be cleaned, instead of having to open up some sort of box. There's not a lot of room in the fenderwell, so any sort of box type thing wouldn't work well anyways, there needed to be a cylindrical shaped box or collector area. The round/ cylinder shape would likely help flow, one would think, rather than have a generic straight sided box.

Even though the fenderwell system works great (much better than open engine bay cold air intakes with cone filters sucking up hot air), i'd imagine that there's still a small percentage of hot air from the engine bay that gets by the separator plate, too (mine isn't totally sealed off), so this would ensure that there's completely cold air going to the intake. Taken from the bumper, it's a legitimate high pressure area, rather than in the hood, which isn't. There's a system made for the 5.0' that looks pretty good that had initially got me thinking of how to route something:

A guy had rigged one up on his SN95 (How to custom build a Ram-Air Intake for your Mustang GT) which had some good results for him, performance-wise (though on these slow 3.8's, the gains obviously wouldn't be as much), though the path of my ram air would have less turns in it, and a non-metal system would likely pick up less ambient heat from the radiator and surrounding area, since the lower part of the tubing is still in close proximity to some engine heat.

 

[BRN2RUN]

I finished the ram air, and it out for a drive and datalogged both before and after. I have more time than sanity, but hey, you know, that may pertain to any of us that decide to work on a 3.8, ha ha. Putting any sort of airbox in the fenderwell was near impossible--nothing other than a tight cylindrical shape would fit in there, and that would likely cut down on airflow quite drastically. The duct funnels right by the cone filter, and you can hear the intake suck in a lot more air, it is noticeably louder. The air still funnels into and congregates in the fenderwell area; perhaps this is "ram area"?

Here was before, with a 70mm Accufab (no ram air) with the intake port matched, and on ported/ polished intakes:

Here's a datalog of a test with the ram air, but with a modified stock throttle body (half shafted, polished, screws shaved down):

I wish that I had logged a stock TB at some point, and could compare that to the Accufab, to see what gains were had and if the manifold was taking advantage of the added airflow. That would provide more accurate results.

Unfortunately, I didn't log the RPM's on the 1st run, and some of the parameters are a bit different on the ram air run (ie: "air mass 1 cyl"), and the intake sensor temp PID seems to be different on the second run. Also, it was a cooler day on the ram air test, which could add to some slightly skewed numbers. If this were on a dyno, in no way would it be SAE corrected; it's just some renegade, on the pavement results (though dynos can be skewed, as well). Take it for what it's worth.

That being said and the fudge factors being out of the way, if you look at the coolant temps and the vehicle speed on the wide open throttle runs which are almost the same (LiveLink gets close, but you're not able to often dial in the exact increment), at the peak numbers, it gained almost a full pound per minute in airflow, and the manifold absolute pressure (inferred MAP), the manifold/ system is at least able to take advantage of the increased pressure, as it shows almost 1.5 kPa in improvement. If the heads/ intakes were rejecting air and the ram air was hurting things, you'd see a drop in kPA.

Edit: the peak kPa shows as 31.875 on both datalogs, but this was a spike early in the WOT run that wasn't at all close to any peak numbers.

 

[Markstang]

the control run looks like it has somewhat valid data. the second run doesn't make much sense at all. I would want it run again.

 

[BRN2RUN]

Did you want me to put the PID's on the graphs? That may help. Here's one with the graphs which shows the whole range of the airflow versus manifold pressure versus speed, which are the main parameters here, I think. The red graph is manifold pressure, the blue line is the airflow in lbs per minute, and the purple line is speed. On the pre-ram air run, I didn't run it out as far in 2nd gear, but in either case, neither was anything comprehensive in a top speed way 1/4 mile dragstrip way, since the intention here was to make a quick comparison of a 1st-2nd gear WOT run (without getting a "display of speed" ticket, of course) .

Pre ram-air, this time at 50.07 mph:

After ram air at 50.03 (this was as close as I could calibrate the datalog info, but note that the MAP numbers and airflow in lbs per minute are still up over the 70mm Accufab):

Notice the first time that the speed (purple line) dips--that's when the car shifts into second. Look at the pre-ram air, and then after. The blue line (airflow in lbs/ minute) is WAY down below it. On the pre-ram air run, that spike in the speed is wheelspin on that particular road/ run. Then look on the ram air run, and the airflow is above the purple line, neck and neck and outdoing that particular race. Once the pressure of the air starts picking up as the car picks up speed, it's clearly doing what it's supposed to. But also, look at the red line......that's our actual power line, since that's what the manifold is actually taking in as air. Notice the drop down from the peak MAP at 31.875 early in the very low rpms isn't as drastic?

This could function as a really crude dyno, I think. Also, look at how much quicker the blue line happens in the powerband in that WOT run. We're seeing the actual effects on what's going on.

I've custom fabbed up a plastic unit to go around the air filter and will log it in the next few days.....this should hopefully add a bit more pressure to the air.

Edit: found this at Vararam's site:

VR-SC1

They mention this:

"its not just peak pressure but the time to peak manifold pressure that we look at, this occurs at shift recovery, down in the meat of the power curve."

Looking at the graphs of the ram air on my car (even without the airbox around it), it's clear that the recovery after the 1-2 shift is startlingly different, as far as being better.

 

[Markstang]

Ah now that looks much better, especially being able to see more of the first gear pull. BUT we don't know how livelink is scaling the y axis of that graph. I could make the first graph appear the same as the second one just by changing the y axis scale and bringing the speed and lb/min lines closer together. We've got two very different values plotted on the same graph and it doesn't exactly specify the scale of each. So I wouldn't concentrate on the relationship of the lb/min to the vehicle speed.

So your ram air yielded a pound per minute of air. I'm afraid that the fact you did the ram air run on a cooler day ruined the experiment. I could completely see that one #/min being because of cooler denser atmosphere. It was a little unclear, did you set up an air box yet or is it just an open pipe pointed at the filter with the rammed air blowing at it? I daresay you might get the same results by simply removing your headlight to let air into the fenderwell.

 

[BRN2RUN]

Very true on the temps......you can see how the fuel temp is cooler on the ram air run, as well. What I'd noticed on datalogging a few times, is that the fuel temps are generally a reflection of the ambient temps, though with some heat in it as the engine gets up to operating temps. Looks like there's a 16 degree difference. So again, this is nothing comprehensive, but offers more data than the good 'ol butt dyno. We've been having really weird weather......great one day, and then much cooler the next (ie: we'd got light snow 2 days ago). I really want to re-do this test on a super hot day.

The Livelink graph was scaled so that it was zoomed all the way in, minus one increment, on both tests.

Also true on direct air flow, and you could likely have the most of the same results by removing the front headlight (though the opening in the metal behind the light looks to be smaller than 3", which is the size of tube that I used, and air would likely get hung up, and/ or it would become a hot air intake at idle if not routed into the fenderwell and sealed off). The otherwise stock fenderwell design obviously just takes air from a general (I assume, low pressure) area. This was just a 3" tube going into the fenderwell, but I'd rigged up an airbox, and should have some results soon. You can definitely hear the added air being taken in, though, and that was happening with the Accufab, but generally lower in the RPM's. You can still hear the air in the upper RPM's with this.

 

[6 Shooter]

Another way to check the differences is to data log with a wideband O2 and watch the A/F numbers to see if the motor ran more lean with more ram air than without. You could do a 2nd gear slow acceleration from a stop and add throttle slowly up to 4000 rpm. Then, use some duct tape and block the entry to the ram air and repeat the 2nd gear slow acceleration from a stop and add throttle slowly up to 4000 rpm. Data logs on a spreadsheet can be trimmed to have equal elapsed times. Then, at the bottom of the spreadsheet, average the separate A/F columns and compare the A/F differences.

BTW, your photos of your porting were amazing and well presented with the before and after results. Magnificent work.

 

[BRN2RUN]

Another way to check the differences is to data log with a wideband O2 and watch the A/F numbers to see if the motor ran more lean with more ram air than without. You could do a 2nd gear slow acceleration from a stop and add throttle slowly up to 4000 rpm. Then, use some duct tape and block the entry to the ram air and repeat the 2nd gear slow acceleration from a stop and add throttle slowly up to 4000 rpm. Data logs on a spreadsheet can be trimmed to have equal elapsed times. Then, at the bottom of the spreadsheet, average the separate A/F columns and compare the A/F differences.

BTW, your photos of your porting were amazing and well presented with the before and after results. Magnificent work.

Thanks! The heads aren't on the car, and I've only partially started the second head. Truthfully, I'm a little afraid of how they will work out.....excited, but scared. I'm aiming to have them on the car the next year, as the T-Bird work took over and it's easier to build power on that one.

You're right on the A/F ratio.....if you look at the ram air datalog PID's in "short term b1/b2" for the short term fuel trims, you can see that it's at .859, Lambda (which I believe works out to 12.7-12.8 ish A/F). But I didn't do the before version. I've noticed that all of the different Bama tunes that I've got are .859 at wide open throttle, but do you think that it would show the difference at partial throttle? One would think so. Thankfully, I'd logged the fuel trims during the whole ram air run, so when I seal the ram air off with my MacGuyvered plastic housing with rubber coupler on it, we'll see what happens.

I was also thinking that a Windstar intake would work much better with a ram air, since you could be getting a straight on ram air setup, and could just run the filter/ ram air over or in the radiator area (but seal it off from the rad to prevent heat from getting in). Someone on here was going to do a Svede Speed custom one, but never got around to posting back on it.....I was hoping that they would.

 

[BRN2RUN]

Here's a datalog of the run with the ram air with a sealed airbox--the numbers are still up over the non-ram air run on the peak numbers, but not as good as the non-sealed run, and the low-mid torque appears to also suffer a small bit, as well. This is at 50.18 mph (as close as I could get it to 50 mph, to compare it to the other runs), when the ram air should be working best. At cruise speeds, the intake was dead quiet, I couldn't hear it taking in air at all like I could on the non-sealed ram air run. In comparison, IAT intake air temp was 66 degrees on the non-ram air run; 64 on this one, just a tad cooler. Though I didn't unfortunately log the "load" parameter on the 1st ram air run, the load on the sealed ram air run is slightly higher on the engine than on the non-ram air run (plus, spark advance is down a bit, likely as a result). The problem in the 1st ram air run is that spark advance wasn't logged (apologies......), but between the non-ram air run and the sealed ram air run, where the intake air temps, coolant (or "ECT") are the same, it's safe to say that the comparison is valid.

The main problem, I think, with the sealed system is that there's no way that you can get any airbox big enough in there to get enough air in a collector area, so the available airflow suffers. It appears that the "ram area" system into the general fenderwell area works pretty well.

There's a few more PIDs/ parameters on this one, and you can see that the barometric pressure is higher than the first Accufab, non-ram air run. The airflow recovery after the 1-2 shift isn't nearly as good as either of those runs, though (the blue line being way below the purple line), so 2nd gear is slower than it should be (though manifold pressure is still higher than the initial run).

For what it's worth, the diy ram air that i'd done on my T-Bird felt like it suffered when it was sealed up, and seemed to gain power when it was just a tube going from the the front bumper to the airbox area.

 

[BRN2RUN]

I did various runs with various configurations (including the ram air setup with the Accufab), and ironically, it was the first ram air run that still had the best numbers. I did a ram air run today with the Accufab:

And here's a comparison with the 1st ram air run a few days ago, but with much hotter outside temps ("ACT" is air charge temp; "inf amb" is inferred ambient temp; "iat" is intake air temperature), and done on the same day as the Accufab ram air. The blue line (air flow) and purple line (vehicle speed) are much better. Again, both of these zoom ins are zoomed in to the second to last increment available on the Livelink:

 

[Markstang]

So what can we conclude from all this?

 

[BRN2RUN]

Long story short conclusion--ram air works! It's not a huge gain, but the graphs show that even on a hot, heatsoaked day, that funneling air into the wheelwell has good results, and that the airflow definetely goes up with vehicle speed and that the manifold pressure is still strong. Also, sealing off the system doesn't work.

Mounted in the front bumper:

In the fenderwell:

 

[Markstang]

Where's your filter located, in the engine bay or through that hole and into the fender well?

 

[BRN2RUN]

My filter is in the through that hole in the fenderwell. The mouth of the ram air, I'd used something called "Stealth Flow" that you can get at hardware stores.....it's meant to hook up to an eavestrough downspout and go under woodchips and things like that, and I cut the mouth section to suit. The flexible pipe was Mole Pipe, which worked out great.....it stays straight as you need, but pulls out and extends as you need it, too. Materials were about $25-$30 bucks.

I may hook up another pipe to the other foglight that runs into the fenderwell filter area and then see what happens with that, too.

 

[BRN2RUN]

Another wrinkle that i'd found in the ram air logging, is the problem of wind: different days create different pressures, and it also depends whether you're facing into the wind or not. Alas, no conclusive data on ram air, once again.

Last year, i'd bought some new OEM Ford valve covers with the improved baffle design, and they were unpainted. Originally, I was thinking of getting the BBK chrome covers, but people had said that they were unbaffled at the PCV area, so this made me decide against them.

As far as I know, other than the BBK's, there's no other options for our engines for valve covers. I painted the OEM Ford new ones red and put them on the car and a few months later, the paint had unfortunately cracked. I had used high temp paint, but the problem was that once oil had got on them (and hot oil, at that, around the PCV and filler cap), that was where the problem was. It was nagging me that the paint had cracked, so I decided to strip all the paint off and go for a brushed aluminum look. It took me a few hours, with a flapper sanding wheel, and they just glisten. I'm really pleased with the results. The most difficult part was getting a smaller wire brush wheel (that is about maybe 3/4" wide) into the areas by the bolt bosses. It was do-able, but was the most time consuming part of it. All in all, it was about 5 hours' work, after polishing them with metal paste and buffing them. I think that the rectangle is screaming out for something--maybe a "Powered by Ford" logo.

 

[BRN2RUN]

Aha--I solved a major problem! With all the work that i'd done, the car just felt slower than it should have been. Even the port and polished upper/ lower intakes felt like they should have been making more power, because there were some sharp and weird, potato shaped casting flaws that undoubtedly held back flow. It's been frustrating, and after looking at some 3.8's doing 0-60 runs on YouTube with similar (and in some cases, less mods), my time seemed to be about the same, if not a tad worse. The car seemed to be faster a couple of years ago, but then I was wondering if I didn't lose some cylinder compression or something.

Some time ago, I had put a 1/2 inch phenolic spacer in between the intakes, with the idea being that it keeps the heat from the lower intake manifold from transferring to the upper, and it also extends the runners so that the long runners yield a bit more torque. I was wondering if making the high end (round, short) runner longer would make upper end power suffer, though. It did keep the upper intake very cool, though the other train of thought is that the EGR is still going to heat the intake up to a certain extent.

Anyways, I took it out on the highway tonight after re-installing the intakes without the spacer on it, and the difference is night and day. It's like the car dropped a few hundred pounds. I'm wondering if there wasn't a slight vacuum leak, because the bolts that came with the spacer didn't extend much into the lower intake manifold, and it really couldn't be torqued down much. I'm assuming that there was also a vacuum leak between the individual runners to other cylinders, so I wonder if some weren't running rich and some running lean.

The other problem that was fixed, is the shifts. On another board, I had a discussion about it and nobody could figure it out--it wasn't trans slip, but the wheels were spinning way more than they should have. The shifts were chirping way too hard, and 2nd gear felt like it had no balls right after that. Now, the shifts are back to quick, crisp and with little tire spin, and the 2nd gear recovery is great. My guess is that the (assumed) vacuum leak was wreaking havoc at WOT with the air/ fuel and load (and not pulling as much air as it should be down the runners and losing velocity).

 

[Phil II]

Perhaps you can find a better length set of fasteners for use with the manifold spacer. I wouldn't expect to find them at most hardware stores, but that would allow you to determine the major diameter and pitch. Bolts with washers could do it. I'd proly dig through some catalogues for studs.

 

[BRN2RUN]

I think that I may look for some longer bolts and then try it again next year (Winter is coming soon). Indeed, the hardware stores didn't have any bolts in a longer length than the ones that I already have.

 

[6 Shooter]

Contact Metric Multi Standard in Irving Texas. You have to provide the thread size and lengths in metric numbers. Bolts come in grades . Higher grades are harder. Just ask. If they do not ship, can get you some.

 

[BRN2RUN]

I think that my goal for next year is to get a '94/ '95 Super Coupe engine and do the swap. My engine is at 206,000 kms (kms, not miles), which is high enough to start second guessing any mods. And i'm kinda running out of things to do on the Mustang, now--camming it would be cool, but guys with the SSM powerpaks usually dyno at about 215-225 rwhp (based on V6Power.net dyno graphs). I was thinking of going with a Mark VIII swap, but after seeing some YouTube videos of guys with various mods (tune, bolt ons, no major headwork), they were sitting at something like 239 rwhp and 250 rwtq, and the engines need to be revved high to get the most out of them.

After working on the 'Bird and figuring out the ways to maximize power through some modifications and tweaks (without getting an actual tune), and getting the SC to be a 5.6-5.9 second 0-60 on an automatic trans in a two ton car, the prospect of putting an SC engine, in a much lighter car--with a stronger transmission with a better 1st gear ratio (and much better traction)--seems like a wild idea. Plus, some things like the BBK shorty headers will already fit the SC block, and (I believe) the valve covers should, as well. The coil pack from the '99 3.8 Mustang engine is the same one as the SC, too. There's other people that have swapped in the SC engine (or put the M90 on a single port), but I can't seem to find much follow up on it in terms of modifications and upgrades like the SC guys have. Generally, it seems like--for the most part--Mustang guys will swap in the SC engine, but not necessarily find out how to make the best power out of it.