Project Shoehorn - M90 into a 4.0L

[Talyn]

That would be the very location I have referenced. I think it was Talyn that pointed me there... Could be wrong about who it was, senile you know...

Yes, there are a couple holes you could use on the manifold side.

2. If you are changing injectors and are using the AEM FIC, the Injector Response Time is a minor but really important thing to set. It has to be set to the response time of whatever injectors are installed. Unfortunately that is typically not a well published value, and not anywhere near the default value of 486 uS.

Its important, but the value is hard to find and AEM's implementation of the latency time is half-assed. There is typically a curve for response time vs. voltage that most fuel ECUs use. They will have a full table for voltage vs. latency. Having one number isn't the best idea. The other thing is that AEM isn't clear on if its the current's injectors time or current minus stock or some other thing like that. I've heard it both ways.

 

[SolarBell]

Its important, but the value is hard to find and AEM's implementation of the latency time is half-assed. There is typically a curve for response time vs. voltage that most fuel ECUs use. They will have a full table for voltage vs. latency. Having one number isn't the best idea. The other thing is that AEM isn't clear on if its the current's injectors time or current minus stock or some other thing like that. I've heard it both ways.

Yeah, I found a good table with a lot of latency versus voltage values for about a hundred different injectors. Unfortunately it doesn't have the Venom injectors on there and Venom is unreachable at this point so I'm using the latency number from OGS's 40# Venom injectors.

I called AEM yesterday and they confirmed to me that the value is supposed to be the latency of the injectors that are currently installed. I can see how it would be confusing though since the PCM is using values for the stock injectors and the FIC doesn't know what those are. If I had all the time in the world I would love to figure out how to build my own FIC-type controller since they all seem to suck in one way or another, but that's way outside my range of time and desire right now.

 

[SolarBell]

Moved the knock sensor over to the manifold side tonight. For future reference, the knock sensor hole is threaded 3/8-16 and is approx. 1/2" deep. The pain was that the Bosch knock sensor I got has a hole sized for M8 and doesn't fit on a 3/8" stud. It's there now though (and I didn't modify the knock sensor, to be clear).

 

[O-Gauge Steamer]

Yah... While I do not remember the actual metric size, it was just a matter of using the drill bit as a ream to open the hole in the sensor. For sure a different part from yours. Glad to hear it is in. Be amazed at how much easier it is going to be to get the timing right under boost conditions.

So, a short note here on ignition timing under boost conditions. The rule of thumb states that you remove one degree of spark advance for every pound of boost. Funny as it may sound, my ignition map almost matches this rule. It is a fair starting point at any rate. Actual values will depend upon the compression ration of the particular engine being fitted with a compressor.

Here is the map I am currently operating with:

[/URL][/IMG]

It may not be perfect, but it does work... The load side is in Pounds per Square Inch, Absolute. That makes the 22 value at the top a reflection of 7.3 (22-14.7) pounds of boost at sea level. If I were to actually go to sea level, this map would require editing to reflect the actual performance. As it is, it is a fair compromise. The AEM F/IC operates in the same fashion as the stock PCM. At ignition on, zero RPM, the F/IC takes a snapshot of it's built in MAP sensor. This establishes the starting point for both the ignition and fuel maps. Once the engine is running, the F/IC goes dynamic.

Ideally, there would be an small bit of electronics that compares the output of a one bar MAP sensor to a preselected value. The circuit would then shift the maps from "A" to "B". The purpose of splitting the maps into "High" and "low" altitude versions is resolution inside of the individual maps. It is obvious that if the pressure range is reduced, the control over the ignition is improved. Unfortunately, I was not able to get the voltages needed so as to design and build this circuit. Here, in Colorado, we have to be prepared to go as high as 14,110'. I pretty routinely take family from out of town up to the top of Pikes Peak. Having a compressor on board lets my Heep scoot right up the hill.

So then, this map accounts for running from near sea level to the top of the Peak plus a small safety margin on the high altitude side.

 

[SolarBell]

Got my baseline dyno numbers today in anticipation of making the switch this Sunday. 138.3 HP, 172,8 TQ Numbers are uncorrected for altitude. Given the HO engine is rated for 190 HP and subtracting 27% for driveline losses, that is a totally reasonable number.

 

[yossarian19]

I'm hot and bothered for the "after" numbers, I can only imagine how you feel

 

[SolarBell]

The shop I took it to specializes in Mustangs so I did feel a little underpowered standing next to a Boss 302 with twin turbos.

 

[SolarBell]

Oh, I forgot to include it the first time since I missed it on the printout - the SAE correction factor was 1.23.

Just so we're clear on factors affecting the numbers (2000 model year):
- Stock headers and downpipe, CA emission package with 2 "pre-cats", and Dynomax 2.5" cat-back muffler and exhaust.
- Spectre cone filter and 3" flexible intake
- No engine mods
- 55mm throttle body
- Air-oil separator (I have noticed it makes a difference, even on NA engine)

 

[SolarBell]

It's in! Well, almost.

I did the final install tonight. Took about 6 hours to swap everything over.

- New intake manifold
- Thermal blanket on headers
- New injectors
- Re-route all vacuum lines
- Finish charge air piping
- Install last bracket
- Move intake air hose

And it all fits. I now have more free space on the drivers side than the passenger side.









I ended up leaving all of the old vacuum lines in place since removing them was more of a hassle than I wanted to deal with tonight. The only things remaining are a new serpentine belt since the one I got is too short by around an inch, and a new clamp for the intake adapter since the 3" one I got is just a shade small.

Hopefully I will be firing it up and start tuning on Monday.

 

[rav]

That is shoehorned all right!, you could not have picked a more apt word. Can't wait for the dyno #'s.

 

[Talyn]

That looks really good.

 

[SolarBell]

And Murphy showed up last night. Here's what happened this afternoon.

 

[kastein]

Wait, what?! It didn't do that itself, did it? What went wrong?

 

[Talyn]

I'm confused. Whats this?

 

[yossarian19]

I'm confused. Whats this?

Disaster.

 

[SolarBell]

It's been a long day and I forgot about the follow-up.

So Monday I fired it up for the first time. It ran, but not great. A/F was way rich so the PCM was running the short term fuel trims as high as 18% - 20% to get it stoichiometric. But that wasn't the bigger deal to me. For some reason at idle I was still at around 12.5 in/hg of vacuum in the charge air piping, which was significantly more vacuum than I expected. Stupid AEM gauge was showing 20 in/hg when I was seeing 12.5 in/hg at the compressor intake manifold, which contributed to my panic. The FIC was showing around 5.5 - 6.0 PSIa. I went through all the numbers and it seemed like something was wrong.

So today in deal with my paranoia and completely rule out blockages and compressor issues I tore the compressor out and checked it for blockages, oil leaks, etc. I didn't find any so I put it all back together. While I was at it I put in a new port for the bypass valve that is much closer to the compressor output to minimize lag. I also cut and inserted rings into the silicone elbows to stiffen them, since under the vacuum they were sucking in really far and I was concerned about them closing off.

I also talked with O-Gauge for a while today and compared pressure numbers with him. It appears mine aren't too far off from his. So other than the fuel maps, it looks to be almost normal. Everything is reassembled now and waiting for my battery to charge since I left my driver's door open overnight. I'm going to work on the fuel maps a bit and try again in the morning.

 

[O-Gauge Steamer]

We did indeed compare values. I started my Heep after connecting the laptop. With the OBDII software and the AEM software running, these are the values tat were recorded.

PCM Side:
MAP = 36 kPa. (5.221 PSIa or 10.632" Hg)
RPM = 860 dropping to 748 after 40 seconds (normal operation).
Spark = 8 degrees. (no alteration via the F/IC)
Long Term Fuel Trim = -10%.
Short Term Fuel Trim = -9%.
Intake Air Temperature = 84ºF = 28.889ºC. The sensor is located in the compressor's intake manifold just under the TB.
Engine Coolant Temperature = 171ºF = 77.222ºC.

F/IC Side:
AFR = 14.8.
Discharge Manifold Pressure = 6.1 PSIa (12.42" Hg or 42.05 kPa)
Discharge Manifold Temperature = 108.43ºF = 42.461ºC.
Injector Duty Cycle = 2%.

Mechanical Manifold Pressure Gauge = 10.5" Hg. (5.157 PSIa or 35.55 kPa) The gauge is an AutoMeter.

Note here that all of the devices bring their respective tolerances to the party. It is for this very reason that I do not rely on any one of them to be a final arbiter. What I do is to average all of the relevant readings when making corrections. Mayhap not the best method but it does have the saving grace of working...

It should also be noted that the Discharge Air Temperature is a calculated value via the F/IC and I am not all that convinced of it's accuracy... A dedicated gauge is on the "list of things to do". Not high on the list, but it is there...

Let's discuss the, IMO, poor Fuel Trims. In an ideal world, the trims would be within +/-5% for the Short Term and +/-3% for the Long Term Trims. Whilst what is running is not ideal neither is it all that bad either...

When I installed this compressor, it was with the idea of running E85 as my primary fuel that drove the injector selection. There are large differences in the Stoichiometric values for these fuels. Gasoline has an AFR of 14.7 whereas E85 the AFR is 9.6. Takes a much larger flow rate injector to make E85 work well.

This is why there are 40 pound injectors in my rig. This is also why the duty cycle is only 2% at idle and is exactly why getting the trims to the ideal is not possible at idle. Once the rig is on the open road, these values drop down to where they belong. Were I to replace them, I would install 34 pound injectors just to have a margin of safety. But then again, I am a bit paranoid.

I spent the better part of a year fiddling with the fuel map in the F/IC in a vain attempt to get the idle trims to the ideal. As it turns out, with a duty cycle of only 2%, the injectors are not operating long enough to make the requisite fine corrections. Given that the trims are good in the higher engine speed ranges coupled with the PCM not being unduly perturbed by the existing trims, I have elected to leave it alone and drive the rig.

As SolarBell can attest, it is a fun rig to drive. My local Jeep Dealer agrees with him on this point. They could not believe how it runs... Turned out that every Line Tech, all of the Service Writers and the Service Manager took it out for a spin after they installed the Transgo Shift kit for me.

All of them agreed that "It does not drive like a lifted Cherokee". Or a stock one for that matter...

On another note, My Wife's Truck had some issues and she had to drive my Heep to work for awhile. She is now nagging me to install a compressor into her '96 Ford Bronco. Last of the Full Size Bronco with the 5.0 Litre engine. This is a very tempting project.

A note on E85. There are engine out there operating on E85 and superchargers that are making some rather incredible amounts of power. E85 burns cooler than gasoline and, as a result, more spark can be introduced without concern for detonation.

 

[Root Moose]

Cool, glad it wasn't a catastrophic failure or something.

Looking forward to seeing it sorted.

 

[SolarBell]

Yeah, it was 85% paranoia and 15% wanting to install a new bypass trigger port. I just fired it up with an updated fuel map and it seems really good. I'll be taking it out for some road tuning this afternoon. More to come.

 

[SolarBell]

Oh yeah - it runs. Still want to do some tweaks, so nothing more to say just yet.