ITB/MS 2 speed density and altitude

[mikesarge]

Went for a nice drive up into the mountains today. Drove like a dream, until I started climbing. By the time I got to the top, it was basically undrivable- SUPER lean when I'd get into the throttle really at all.

Lean is not what you'd expect at altitude, but because my VE table looks like this:




I'm assuming at altitude, what was the throttle position at ~93kpa at sea level, was a lot lower in terms of TPS % than up high, where WOT was only giving me about ~93kpa.

Does that make sense? i.e., we're fueling just a little bit down this steep map (which is WAY less fuel), despite the TPS being closer to 100%, and the reduction is disproportionate to the actual loss in air density.

That's basically what was happening: around 2k-4k, and between about 88kpa-95kpa, I was spiking SUPER lean. I had my laptop with me, so I added some fuel to those bins, then the laptop died. On the way back down the hill, the fuelling came back to normal where I didn't screw with it, and was SUPER rich where I did. So, I fixed it when I got home and now we're back in business, as long as we're at sea level.

So, does anyone else run speed density and drive up mountains? what do you do? %baro? do you have to add a second pressure sensor? calibrating barometric correction seems to take me in the wrong direction: I need to stay in the right place on my map, not take fuel away from the top.

Maybe like a reverse ITB mode, where you use %tps at the top of the map, rather than the bottom? that seems ridiculously complicated, but I guess I'm open to whatever!

any help is greatly appreciated!

Edit: here’s my proposed solution: change from speed density to %baro, then figure out how much fuel to pull on baro correction, once I’m up high. I think I’ll still need to add a constant baro sensor, and the diyautotune harness/mega manual do not make it totally clear how to do this. More reading!

[phelix]

I would have thought a barometric sensor was the right answer?

[mikesarge]

Originally Posted by phelix

I would have thought a barometric sensor was the right answer?

‘Barometric sensor’ but correlative to what?

My tune currently uses speed density to calculate fuel, so kpa is already there: the engine knows what baro is, because it samples it at start up, and it’s reading manifold absolute pressure all of the time. That doesn’t compensate for the issues with the fuel map, where WOT is now down at 93kpa, and my engine is running extremely lean.

[phelix]

a sensor measuring ambient air pressure.

let me try to explain… as you go up in altitude the air density decreases. if you’re only using a MAP sensor to determine
fuelling the sensor will give a lower pressure reading at a given engine speed and load so the fuelling will be less, i.e. lean. you really need a second input, either TPS% or ambient air pressure. without that you will have issues at altitude and could even see running issues on high or low pressure days, depending what the ambient air pressure was on the day the car was mapped.

[mikesarge]

Yeah I understand all of that, I think we’re saying the same thing to each other.

My issue is basically one of hardware and software: in tunerstudio, you can sample baro at startup (using MAP, with engine off) but if you’re fueling based on kPA, that initial value is meaningless without baro calibration. If you use barometric calibration values, it adjusts the entire map on a percentage basis. This doesn’t fix my issue, where the percent the value is off varies based on where on the map you’re measuring, because WOT at 2k’ above sea level is 92kpa, and my map is giving way too little fuel for 85kpa at 2k’ above sea level.

Similarly, if you’re fueling using %baro, you only get the baro at the time/place you started the engine. I think where this leaves me , is I have to figure out how to add a constant barometric pressure sensor to my ecu somehow, so that I can run fueling based on %baro, with live baro sampling, then create a baro calibration value so that it takes fuel out up high, rather than my current condition where I go lean up high running pure speed density.

[shamrok]

Curious what you find out.

I just moved from sea level to ~7k feet, and the "sample baro at startup" option did not work for me at all, it would not actually read ambient barometric pressure and after asking around and reading, it seems you really need a second MAP sensor to really use the baro curve.

I ended up manually setting the default baro under General Settings to 80 kPA, which aligns with my altitude and it seemed to work. Having to manually set the default baro versus it sampling at startup is annoying and I'll eventually add another MAP sensor to get it working correctly, but like you have discovered, the documentation is quite weak regarding how to do so.

[mikedsilva]

Interesting thread.. I never understood this issue. Someone tried to explain it to me saying the ecu (in my case AEM) made all the calcs in the background and adjusted accordingly.. but it didnt have a barometric sensor so I don't really know how it knew...

[Showdown]

I’d you head over to the msextra forum this topic is akin to “what oil should I use” over here. That is to say, there are dozens of posts and discussions about it.

The solution is pretty simple even if you need to do some soldering.

https://www.diyautotune.com/product/mapdaddy-4-bar-map-sensor-with-barometric-correction/

https://www.msextra.com/doc/ms2extra/MS2-Extra_Hardware.htm#conbaro

[mikesarge]

Originally Posted by Showdown

I’d you head over to the msextra forum this topic is akin to “what oil should I use” over here. That is to say, there are dozens of posts and discussions about it.

The solution is pretty simple even if you need to do some soldering.

https://www.diyautotune.com/product/mapdaddy-4-bar-map-sensor-with-barometric-correction/

https://www.msextra.com/doc/ms2extra/MS2-Extra_Hardware.htm#conbaro

Yeah the posts on msextra left me with a ton more questions. Good analogy to oil threads, it really is like that over there, I had no idea it’d be that controversial. Anyway, after reading for a few hours, I think the mapdaddy (or something like it) is probably the ticket. My concern with mapdaddy is resolution, but I guess I can continue to use the stock sensor for map, and the 4-bar for baro (since it’s going to have way lower speed changes, doesn’t need as high res down low)

The other thing, is I’d be happy to use a GM 1 or 2bar sensor, but the megamanual is totally silent on what I’d need to do to my board to add it. The diyautotune site sells a Gm sensor, but no instructions on how to modify their board to wire it in? Mapdaddy does have instructions, so maybe I kind of crib from those? I can’t be the first to do this on an NA engine!

I’ll leave this thread out to soak for a few more days and see if anyone else has solutions that work.

Thanks!

[mikesarge]

Originally Posted by mikedsilva

Interesting thread.. I never understood this issue. Someone tried to explain it to me saying the ecu (in my case AEM) made all the calcs in the background and adjusted accordingly.. but it didnt have a barometric sensor so I don't really know how it knew...

I think some folks give EGO a bunch of authority (like 20+%). An AEM ecu probably works fast enough to do this, but in my experience, the megasquirt does not, especially with ITBs, where the fueling is really peaky. I’m considering trying this route until I get a barometric sensor installed though

[Showdown]

As I understand it, the MAP Daddy has a 4 bar for standard sensing and VE control, and a 1 bar for baro correction so you won’t lose resolution for engine management. And as baro changes will (should, hopefully; you’re not an airplane) be slow and relatively linear A 1-bar will be fine.

This update is on my list for my car eventually

[mikesarge]

Installed the mapdaddy this evening and tested it to verify both sensors are calibrated and working. Looks good to me. I’ve changed my fuel load from speed density to %baro, changed min kpa to 70 on the baro sensor, and put in a conservative guess for a baro correction curve based on a bunch of research.

I’m unfortunately slammed this week, but at least this Saturday’s weather is looking good for a hill climb so that’s what’s on the menu for the weekend. I’ve got a nice slow steady climb to 2k, then a pretty quick progression to 6k. I’ll (of course) bring the laptop and try to tune for the 0-6k correction range to the best of my abilities. I think this will test the backpressure (more fuel as you climb) versus thin air (less air so less fuel as you climb) theories for altitude change/enrichment pretty well, so I’m excited to see how it goes. My guess is less air, so less fuel, as long as you’re on a %baro tune. Maybe it’ll be a little of both and I’ll get completely screwed, who knows!