codrus

Yeah, the new FM cats are metal core instead of ceramic. It plus the 3" made a huge difference in back pressure when I measured it right after I bought it (I've lost the data for it, unfortunately).

So today I started by swapping out the warped front rotors on the Audi, and discovered this:



WTF, this tire only has about 5K miles on it?!?! It's through the tread, through the cords, through the steel, and into the inner carcass, I'm amazed it's still holding air! The passenger side front is like that too, and it's a really weird wear pattern. I think what's going on is that the tires are deforming under hard cornering and rubbing on the upright, although why the RE11s are doing that and none of the previous 5 or 6 models of tire I've had on the car did, I have no idea.

So then I moved on to the Miata. I've got a walbro pump in the car, which fixed the high-power fuel droop I'd been seeing before with the 2560 and the RC550s, but its performance with the ID1000s and 2863 was unknown, so I ordered some parts to monitor the pressure:




Mounted it up (I actually wound up using rubber line and brass hose barbs temporarily -- the fancy AN stuff will go in later) and went out to do some logging. It looks like I'm seeing 68-69 psi at idle/cruise, dropping to 61 at max duty cycle. Anyone know if this is normal/expected? I'd hoped to see it be flat all the time...



Looks like I'm at 65% duty cycle on my ID1000s, glad I didn't buy the 750s!

I also added the 3 psi inner spring to the EWG, for a nominal 10 (7+3) psi base pressure. It turns out to actually deliver closer to 9, but OK. This does seem to help with tuning EBC for 280 kpa, although we didn't get a chance to do a full tuning run. Picked up new bolts/nuts for the V-band flanges from mcmaster, the ones that came with the EWG do NOT appear to like the heat, and had seized, I broke one of them getting them out.





--Ian

codrus

Also, the Miata is leaking coolant around the cap on the TSE rad (anyone had one of these caps go bad?) and has a puddle of something oily in the belly pan right under the oil cooler & steering rack. I dunno if it's oil or power steering fluid yet. I guess I'm driving the truck to work for a few days!

--Ian

sixshooter

iTrader: (12)

Does that Audi have one of their double miniature dog bone setups at the top/middle of the strut?

codrus

I'm not sure what you mean by "dog bones".

It's a shock, not a strut, so there are four control arms. The upright curves around the inboard edge of the tire and the upper control arms attach to the top above the tire tread. This is a fancy billet/etc aftermarket setup, but the geometry is the same:



So in normal operation, the upright is pretty close to the inboard sidewall of the tire. I'm thinking the most likely answer is that the RE11s are deflecting more than previous tires were, and that's causing it to rub.

--Ian

sixshooter

iTrader: (12)

I don't know much about those cars but I think I've heard about funny issues with those upper arms developing play and causing strange alignment variations through its range of motion. It could be worth a look, and may be worth a little supplemental reading if you are getting strange tire wear.

patsmx5

iTrader: (16)

Fuel pressure dropping is from either not enough fuel pump, or wiring to pump is weak. Likely the wiring. I have measured voltage drop on my 99 with a Walbro 255HP and it's terrible.

codrus

The bushings in the upper arms do wear out, but the ones on this car were replaced about 18 months ago, so I don't *think* that's the cause. I'm also not sure how excessive toe would have caused a concave wear pattern like that, but...

Gonna call Tire Rack tomorrow and ask if they've ever seen anything like this. Two of the tires are trash, so rather than order 2 RE11s and get them shaved to match the rears (current at around 6/32 tread) I think I'll just be getting four new ones for it anyway.

So with the Miata and the Audi down, I drove the truck around today, and the AC in it failed. Not a good weekend!

--Ian

codrus

Right, but I'm kind of surprised to see 68-69 psi at idle, from the shop manual the spec is 53-61. I've heard of high flow pumps overwhelming the stock regulator -- perhaps it's 68 at idle because of that and down to 61 at high duty cycle because that's where it's supposed to be?

--Ian

18psi

iTrader: (76)

That seems likely. This is the Wally 255 HP right?

patsmx5

iTrader: (16)

Oh I thought you had aftermarket regulator. Yeah you're overwhelming it then at idle and it's going back to normal under load. That will make tuning a pain BTW, as fuel PSI will vary on you changing your fueling at light loads.

codrus

It's the Walbro 255, yeah. I bought it right before everyone decided the Walbro sucked and the DW was the pump to get instead.

I guess I need to go do some research on what aftermarket regulators fit the NB non-return system.

--Ian

patsmx5

iTrader: (16)

haha not sure why they suck, mine's been a champ! It's a bit noisy but otherwise solid. Had mine since 2008 or so, courtesy of Rafa on here.

codrus

I ordered half of an FM big fuel kit, it puts an AEM adjustable regulator where the factory fuel filter goes. Half because I already have the pump, the rail, and the AN lines to feed the rail, so all I need is the regulator, new filter, and the -6 lines.

I'm hoping this will go a long way to fixing my unrepeatable idle characteristics!

Also, I found the Audi tire rub, see the shiny spots on the wiring harness shield and the spring perch?




I think it's the spring perch that's doing the real damage, the tire is effective shaving itself against a curved piece of sheet metal. That totally explains the concave, scooped-out shape to the wear.

As for why it's rubbing on there, I measured the RE11s that are on the car and compared them to the 12-year-old OEM Continental that's mounted on the spare. They are both labelled 235/40R18, but the Bridgestone is 3/4 inch wider, and also has a much more square tread profile. This combined with the wheel being closer to the perch by virtue of more camber makes it rub.

As for the Walbros sucking, I have the vague impression that there was a run of defective units that just spontaneously died. I remember watching Savington rip one out of his car in the paddock at Laguna once when it died during a Miatas @ MRLS event.

--Ian

18psi

iTrader: (76)

I think the main beef is that it's an older unit, not officially made for ethanol compatibility, and loud. Otherwise nothing wrong with them.

So with the FM setup are you converting to return-style? Or just using a stronger regulator? Either way it should be an improvement

codrus

The stock NB system puts the regulator in the tank, but the line that feeds the regulator actually goes out of the tank, to the fuel filter, and then comes back. The FM BFK removes the regulator in the tank and installs the AEM regulator where the filter would normally be. So it's sort of halfway between a return and non-return system.

--Ian

EO2K

iTrader: (37)

Is it actually boost/vacuum referenced? If not, I'm not sure what that gains you other than adjustability for base pressure.

Please take lots of pics though, because I'm curious what kind of lines and adapters they are using and where they are routed.

codrus

It can be boost/vacuum referenced if I run a vacuum line back to it, although FM doesn't typically set it up this way.

The problem I'm trying to solve is that the necessary idle fuel VE map is not always the same. I can tune the car perfectly, then park it, and the next day it's idling like crap. I haven't completely nailed down why yet, but I think it's temperature related. I have two theories as to why this might be, and the new regulator will help with both of them:

1) It should be capable of actually regulating the Walbro at idle, which the stock regulator isn't. With the stock regulator, the fuel pressure at idle is effectively unregulated, which means it varies with the performance of the pump. Things like temperature and voltage will vary that performance, and lacking proper regulation this means that they show up as variances in the fuel pressure. A perfectly tuned idle fuel map is going to be lean if the idle pressure drops from 69 to 65, for example.

2) The injector dead time varies with the temperature of the the injector, I think because the resistance in the copper in the coil goes up as the temperature goes up. This means it takes longer to build up the magnetic field and open the injector, thus longer dead times. With high impedance, 1000 cc/min injectors, even tiny changes in the dead times can result in massive changes in the amount of fuel delivered. The regulator won't directly address this problem, but if I manifold-reference it (instead of atmosphere referenced like stock), then I can lower the static pressure at idle while keeping it at the necessary level in boost. This means I can lengthen the injector pulses at idle, thus reducing the effect of a change in dead time.

And yes, I'll take a bunch of pictures. FM has some in the instruction PDF: https://www.flyinmiata.com/support/i...g_fuel_kit.pdf

--Ian

EO2K

iTrader: (37)

Awesome, thanks Ian. I know you'll document the hell out of it and I appreciate the commentary. It all makes perfect sense, and I know you'll provide solid data after implementation.

I never noticed the FM instructions recommend putting the damper on the bottom of the fuel rail. My M-Tuned has the boss for this and I understand in theroy want it does, perhaps I'll add it back into the mix before I'm done.

codrus

So a couple weeks ago I ordered some parts from FM to address the fuel pressure problem. They sell a "Big (Flex) Fuel Kit", which is basically a DW pump, AEM FPR, GM flex fuel sensor, fuel rain, and AN lines and fittings to connect everything. Well, I've already got a pump (a Walbro rather than a DW), an FM fuel rail (the old purple one rather than the current model, but it works well) and I'm not running E85 just yet so I don't want to go with the flex fuel sensor. So I basically ordered about a third of the parts for the real BFFK:



Here's the AEM FPR with the fittings on it. The 90 degree is the inlet, the straight fitting is the return to the tank, and the big blue thing is an inline filter. They have straight threads (not NPT) and use aluminum crush washers instead of O-rings. The Earl's filter actually has male AN fittings on both ends, but one of them just screws into the female threads in the AEM with the flare not being used. The open port is for the pressure gauge.




Since you need to remove the stock regulator, I had to pull the pump out again. The Walbro is still there, but the pipes are interesting purple color -- dunno why. Zip ties apparently turn brown when immersed in gasoline.



Regulator gone:



Pump back in the tank, with the OEM quick-disconnect-to-AN fittings installed. These are the fancy ones with the aluminum screw lock, rather than the plastic ones.



The stock fuel filter comes out, and this bracket gets bolted in in place of it. Apparently the tolerances on the laser (?) cut bracket are very tight, the bolts needed some coaxing to go through it. Also, the bolts that FM supplied to attach the AEM to the bracket were way too long, dunno why. I scrounged up some replacements that were shorter.



AEM bolted in place. FM supplies two pre-made 24" AN -6 hoses for the tank connections, both with a straight fitting on one end and a 90 degree on the other. The return line really wants a 90 degree on the top where it goes into the tank, and that lines up nicely with the straight fitting pointing up from the AEM. The pump output line wants a straight fitting at the top, but having the two 90s at the bottom is a bit inconvenient. I'll probably buy another straight -6 fitting and replace the swivel here.



Fuel lines at the top:



Hooking up the feed line to the front. It's a 6 or 7 foot AN -6 line, again with a straight on one end and a right angle on the other. Since braided stainless likes to chew through everything, I generally like to protect it with some kind of covering. I had a bunch of this large diameter plastic wiring loom sitting around and it turns out to be exactly the right size to cover the stainless line, so on it went.



I also decided to route it up along the PPF instead of where the hard line had been:




Up top we added an inline AN -6 tee to feed the MS3's fuel pressure sensor. That's still a rubber hose for now -- I have the parts to make an AN line for it, but it was getting late.




Hooked it all up, did a leak test. We'd forgotten to tighten one of the AN lines, it made a big puddle. Once that was fixed everything seemed to be good, we set the pressure to 60 psi, the car started and idled fine. For now it's still non-manifold referenced. I'm going to see how the car does with that, if it fixes the inconsistent idle tuning problems that I've been seeing, then I'll probably stick with it that way. If it continues to display temperature-based idle fueling inconsistencies (thus suggesting that the dead time is varying with temperature), then i'll probably add the manifold reference to effectively increase the idle pulse length so that changes in dead time have less effect.

It was pretty late when that was all done, so we didn't drive the car yet, but it all seems to work so far. The fuel pump is noticeably louder now, perhaps the AN lines transmit more noise, or having the AEM hard-mounted to the chassis does that, I dunno.

--Ian

patsmx5

iTrader: (16)

I recommend go ahead and do the vac reference. Better idle, and more fuel at high load since you can maintain your fuel pressure delta across the injectors. Without it your injectors get "smaller" as you add boost, which sucks, and also makes atomization worse too. There's no downside to running the reference, and a few upsides. You just have to run a vacuum hose back there.