MS3X and MAP sensors question

[chrisbalich]

I'm nearing the final stages of my MS3X and ITB installation on my SC.
I'm doing my best to make the install as clean as possible and I've come to a bit of a crossroads. I intend to run dual MAP sensors to provide altitude correction and I'd like some guidance.

Would it be better to:
Run a GM 1bar on the engine with the internal MAP for baro?
or
Buy and install a MAPDaddy so there are no wires and just a long vac line?

I get the feeling I'm not totally overthinking this as a ton of MS3X users run dual MAP sensors. I do 100% intend to drive from my sea level home to at least 6000' (friends in Denver).

Any input here would be very much appreciated.
(pic because threads without pics are no fun.)

[spuggy]

I run a GM 3 bar sensor plumbed to the manifold, and a 1.1 bar sensor next to it for baro ref.

A 4 bar sensor will be less accurate around 1 bar; not by much, but certainly a couple of %. That's intended for use with forced induction, obviously.

If you've got an on-board sensor on the MS, I'd probably use that for baro and one on the motor for the manifold, because I'd rather have wires than long vacuum pipe runs (for no especially good reason). And it's pretty rare conditions where atmo actually reads over 1 bar anyway.

[chrisbalich]

Originally Posted by spuggy

I run a GM 3 bar sensor plumbed to the manifold, and a 1.1 bar sensor next to it for baro ref.

A 4 bar sensor will be less accurate around 1 bar; not by much, but certainly a couple of %. That's intended for use with forced induction, obviously.

If you've got an on-board sensor on the MS, I'd probably use that for baro and one on the motor for the manifold, because I'd rather have wires than long vacuum pipe runs (for no especially good reason). And it's pretty rare conditions where atmo actually reads over 1 bar anyway.

That's pretty much what I'm thinking. Thanks for confirming my suspicions.

[spuggy]

Originally Posted by chrisbalich

That's pretty much what I'm thinking. Thanks for confirming my suspicions.

Sure, np. You did fit a CHT, right? (lot easier to do it now)...

Trying to go off oil temp is a crock/myth. I have a chaincase temp sensor that is almost completely useless (still rising long after oil tank is up to temp). And removing/replacing the rear tin to fit a CHT with the motor in the car is a little fiddly...

Oh - and if i had everything out, i'd fit a flex fuel sensor in the blink of an eye. Makes no sense not to.

[chrisbalich]

Originally Posted by spuggy

Sure, np. You did fit a CHT, right? (lot easier to do it now)...

Trying to go off oil temp is a crock/myth. I have a chaincase temp sensor that is almost completely useless (still rising long after oil tank is up to temp). And removing/replacing the rear tin to fit a CHT with the motor in the car is a little fiddly...

Al Kosmal sent me a single-wire CHT that he uses when I bought my kit. I've got it cinched in behind a chain box nut right in front of the oil pressure sender.



Since I've got your attention and you are MS experienced....

Any recommendations for wiring the fuel pump? I've isolated the brown/red wire from the CIS harness and I've located the fuel pump output from MS3....but it's my understanding the MS3 output is a ground and the CIS input is expecting voltage? Should I just wire in a relay to fire the stock relay? Or is there a better way?

[jpnovak]

I would use a 2bar MAP sensor for your main EFI reading. Better resolution than the 3 or 4 bar but you probably wouldn't notice it. I do not run 1 bar sensors. I have had problems in summer heat of TX where high pressure can cause the 1bar sensors to reach the upper limit. The problem is that they overflow back to near zero and that means full fuel cut. Watch out on the GM sensors. I had a 2bar PN sensor that was only a 1 bar sensor internally. That took a while to figure out the issue with WOT fuel cut.

I typically mount the MAP sensor right next to the vacuum manifold and then run wires in the harness with the rest of the EFI. You can use the internal 5Vref that is also shared for TPS.

The baro correction sensor can be a 1 bar. I would solder a 2 bar to the board and use this as the baro correction. Direct, wire free installation and no hoses required.

for fuel pump... run your own wiring. Don't try to use the factory wiring. 12V switched (and fused) to the pump. Relay ground through the MS3. The fuel pump connection is ground switched. This will give you 5s (or 10s configurable) fuel pump prime pulse and then continuous operation if rpm signal is detected.

[chrisbalich]

Originally Posted by jpnovak

I would use a 2bar MAP sensor for your main EFI reading. Better resolution than the 3 or 4 bar but you probably wouldn't notice it. I do not run 1 bar sensors. I have had problems in summer heat of TX where high pressure can cause the 1bar sensors to reach the upper limit. The problem is that they overflow back to near zero and that means full fuel cut. Watch out on the GM sensors. I had a 2bar PN sensor that was only a 1 bar sensor internally. That took a while to figure out the issue with WOT fuel cut.

I typically mount the MAP sensor right next to the vacuum manifold and then run wires in the harness with the rest of the EFI. You can use the internal 5Vref that is also shared for TPS.

The baro correction sensor can be a 1 bar. I would solder a 2 bar to the board and use this as the baro correction. Direct, wire free installation and no hoses required.

for fuel pump... run your own wiring. Don't try to use the factory wiring. 12V switched (and fused) to the pump. Relay ground through the MS3. The fuel pump connection is ground switched. This will give you 5s (or 10s configurable) fuel pump prime pulse and then continuous operation if rpm signal is detected.

Jamie, you're the man.
I'll get cracking on this all straight away.
Thanks so much.

[spuggy]

Originally Posted by chrisbalich

Al Kosmal sent me a single-wire CHT that he uses when I bought my kit. I've got it cinched in behind a chain box nut right in front of the oil pressure sender.



Since I've got your attention and you are MS experienced....

Any recommendations for wiring the fuel pump? I've isolated the brown/red wire from the CIS harness and I've located the fuel pump output from MS3....but it's my understanding the MS3 output is a ground and the CIS input is expecting voltage? Should I just wire in a relay to fire the stock relay? Or is there a better way?

Actually, I've got a Motec, but I've had to get a little, uh, intimate with it...

I had the same thing; already had the front relay wired to the ECU (I didn't do that bit). When enabled fuel pump control in the ECU, pump ran continuously (same as before).

With Motec, just reverse the polarity - and it stuffs a control voltage down that pin instead. Flip one cell, and got the 2 second prime and pump off until motor running.

if you don't have that feature in the MS - eg it'll only control a sink to ground, - then, yes, you should probably provide switched 12V to the other control pin on the factory relay and let the ECU provide ground to control the pump.

There is also, on my car (a '77), an entirely unused wire in the factory harness that runs from the 14-pin connector and terminates in a spade by the front relay/fuse panel. I located it and unmade the connector, only to dig behind the fuse panel and realize the relay was already connected to the Motec via a new milspec wire routed through the tunnel... So used that instead.

[Peter M]

Just a question from someone not familiar with ITB load modelling.

Are you using a combination of alpha n and speed density for your fuel modelling and this is the reason you need a second MAP sensor for ambient pressure correction when you are in the load range where you are primarily relying on alpha n?

Thanks
Peter

[jpnovak]

This is purely a continuously adjusting altitude compensation. Kinda like PSIa vs PSIg...

The other algorithm is to use an initial MAP reading when the key is first turned on. In this case the altitude compensation is set during the first startup and maintains that reference for the duration of the drive until the EFI is restarted at which point it saves a new reference.

The TPS vs MAP vs blend fueling algorithm (load calculation) is another subject.

[Peter M]

So Jamie, is Chris is using solely speed density for his fuel modelling?

If that's the case why do you need to compensate for ambient pressure using a second MAP sensor?

From the engine's point of view isn't the absolute pressure in the intake manifold all that's needed to model the VE from?

Just asking to understand! :-)

[jpnovak]

I am not sure how Chris has his EFI configured. Yes, the standard speed-density calculation "should" be self correcting as long as the air pressure vs air density calculation (and vs T) curve is correctly set. But these corrections are only as good as the input and place of sensor of measurement.

I admit I have never installed a second MAP sensor for correction. I see this as a compensation for boosted applications rather than NA installs.

[spuggy]

I've never looked into how VE or load is calculated on an N/A application - because not applicable. Dunno how that works...

With forced induction, typically use MAP as the ECU's source of load & VE, and configuring it to use MAP/BARO instead is the same thing but with compensation for lower atmospheric pressure.

I've driven some shallow mountain passes without, and I don't plan to drive Pike's Peak anytime soon. But it's nice to know that should you drive up high enough to affect fuel trim (and 6000 ft would certainly be high enough to change things noticeably I think), it'll Just Work...

In Denver, "normal" atmospheric pressure is 12 PSI instead of 14.7.

Random quotes from the Motec forums that may be helpful to the OP:

The thing you'll find with change in Ambient Pressure that often gets overlooked is that ignition timing will affected due to the pressure differential helping to scavenge the exhaust gases better at higher altitude [lower Ambient Pressure].

You can tune a NA engine based on TP but with a MAP comp and connect the MAP sensor to the manifold. This normally works well but there are exceptions where a stable MAP reading is difficult to obtain, for example on a multi throttle setup or an engine with a lot of cam overlap (or in the case of a Rotary a lot of "port overlap"). The advantage of tuning this way is that the ECU will respond bettter to load changes at low or closed throttle and/or if an idle control valve is being used. I would still trigger the accel enrichment from TP but you will find that you do not need as much. This setup will also do a reasonable job of compensating for changes in altitude. You can have a seperate baro compensation though if you find that there are changes in mixture that the MAP compensation is not correcting.

If you tune your engine based on TP with MAP comp the fuel table will be flatter because the MAP compensation is also contributing to the calculation of the fuel requirement. The change in MAP when you open the throttle also helps provide more acurate fuelling during transient opperation and can reduce the requirement for Accel Fuel.

...

A seperate Baro sensor can be used and is a good idea if the engine is opperating at different altitudes. Normally Baro compensation is the standard "double the air double the fuel" formula but if it is used in conjunction with a MAP sensor it can be used to compensate for changes in VE as a function of ambient air pressure. For example if you run an engine at altitude with 80 KPA this will have an effect on the exhaust back pressure which can change the VE even on a turbo engine. In other words if you have a turbo engine runnig with 1 Bar of boost at sea level the fuelling requirements will be different if you run the same engine at 1 Bar of boost at the top of a mountain where the air pressure is lower. The pressure in the intake manifold is the same but the exhaust pressure is not and this affects the efficiency of the engine. On an NA engine this also changes, partly because of the lower air pressure in the intake but it also effectively changes the "tuned length" of your manifolds.

[Peter M]

Thanks Jamie and Spuggy.

[shamrok]

Chris, are you planning to run Speed Density or ITB mode? Just another data point, I was running a single 1 bar MAP sensor (Microsquirt, so no internal MAP) with the intent of doing ITB mode, but the car is running so well on TPS-only that I've ditched that idea. I'd intended to wire up a 2nd MAP sensor and do the baro correction thing later...but since I already have a MAP wired up, I'll just retask it, as I'm not using MAP at all, beyond the initial baro reading.

Agreed with Jamie about the fuel pump, wiring it up to match the MS's documentation and its expected relay behavior saved me from further headaches when I was troubleshooting other teething issues. I ended up just running my own wiring directly to the fuel pump and didn't bother trying to tap into any of its stock wiring.

[chrisbalich]

Hey team, been a busy week at work so I'm just now getting caught up here on Pelican.

I have a base map in my MS3X from Al, but plan to run the blended ITB algorithm as my end-goal. I live at sea-level, but make annual trips to the mountains in NC/TN and the Poconos...and I have at least one Colorado trip (Crested Butte) planned for the next 10 months. The NC/TN trips have long drives where we gain 2-3000' of elevation and the temps will swing from the 40s when we leave, to 70 as the day warms and down into the 30s at elevation and then back to camp in the 60s/70s. I don't want to have to stop every hour or two and restart my car so the EFI can re-calibrate for temp/altitude swings. I want the car to be as smooth and reliable as something running a modern Tier 1 management system. In my mind, that means I provide as many inputs as are feasible and then tune accordingly.

I'll be first to admit I'm not nearly as well versed as Jamie or Spuggy or Al. So I'm grateful for everyone's input and questions. This is my first 911 and second MS setup. (The other was MS2 on an m20 powered e30.) I've got plenty to learn and am reasonable enough to defer to experts when possible.

My [open element, GM] IAT sensor is mounted into the baseplate of my rain hat, the MAP will get its source from a vac manifold tied into all 6 throttles, and I'm running DC19 cams (just a trifle hotter than 964). I expect I'll have no trouble getting a smooth MAP signal and the IAT should read exactly what the engine is taking in.

I have wired my EFI to run sequential ignition and injection so dialing in the tune should be pretty straightforward once I get it running.

I am 100% thankful for, and open to, suggestions.

[spuggy]

Well, good luck! i'm sure it'll be very rewarding, once you get it to the seemingly-endless "just tweaking it slightly" stage...

One bit of (hopefully) sage advice; the motor wants what it wants. it'll tell you when you're going in the wrong direction, if you listen to it. If this doesn't fit your idea of what the motor should want, either your idea is wrong, or something else isn't quite as you think it is...