Part Throttle Turbo Tuning

While taking baseline data on my SR I encountered an interesting anomaly during part throttle ramp runs that had me stumped for a while.  I was noticing some very rich wideband readings that didn't seem to match the fuel map.

This is the data from a previous run.  As you can see at only 5psi it is running down into the 9's.  At first I thought it may be compensating for knock, but the logs show that the timing is dead on to the timing map demonstrating no evidence of knock compensation.  So I was figuring it must be an enrichment table adding extra fuel but have no idea which one would do that.  Recently I did a few ramp runs at different throttle openings and found some rich areas that are the same, but others were much more reasonable.

With the extra data the picture starts to become clearer.  There is a pattern with lower throttle openings at this RPM and load the mixtures get very rich.  Also you can see that the map sensor is reading much lower psi at the same load column.  So what is going on?  The theory that I have come to is in this rpm and load area the turbo is starting to come up to speed and is flowing a lot of air (this is confirmed by high MAF readings), but the low throttle opening is impeding the flow into the manifold as indicated by the low psi. (pressure drop across the throttle body)  Basically the ecu is using the wrong load columns because the maf is reading a higher flow than what is actually making it into the cylinders.  When the throttle opens farther the maf reading aligns with the map readings and produces more reasonable afr.  Hopefully I am making sense here.

Has anyone else noticed this on boost part throttle behavior with a maf based ecu?

I'm not sure how to approach tuning tuning part throttle under boost.  I am leary of leaning out the fuel map because then it will run lean at large throttle openings in those load areas.  Any suggestions on how to address this tuning challenge?

Just talking off the top of my head here:  More background?  What's your injector setup like?  My thought is maybe your secondary injectors are generously sized and coming in around 4000 RPM.

Your post references both MAP and MAF sensors.  Can you elaborate?

Been a long time since I've fiddled with the Haltech on the RX-7 and I've probably forgotten 90% of what I once knew, but I'll be happy to try to keep this thread active. 

Stock injectors.  MAF is used for load.  I added a map sensor just to be able to datalog boost pressure for reference.

There is usually a TPS* and MAP* enrichment parameter. On my boosted projects, it has taken some fidding to prevent the MAP* (map dot or map rate of change) from adding a bunch of fuel if I'm at steady throttle but boost is ramping quickly. without seeing the time domain data, my guess is that you have too much MAP* impact and need to move some of that over to the TPS* enrichment.

In reply to gearheadE30 :

Came here to say exactly this - review the acceleration enrichment settings.  Might be using some sort of MAF rate of change calculation, since it's a MAF based system?

It's been a very long time since I've touched the tune, but I believe I finally gave up on MAP-based enrichment on my turbo escort and switched over to a basic TPS-based accel enrichment.

One thing I normally do when I am tuning a main table is filter my data so any accel based enrichments aren't included. Not sure if that's possible with your software but if you can output to csv megalog viewer does a good job in the histogram view once you add the filters (such as TPSDOT > 100%/s etc or if there is an accel flag)

Sounds like too much MAP based enrichment to me, too.

The ecu has no input for MAP.  I have a GM map sensor hooked up to the data logger purely for data collection.  The ECU bases load off of MAF with enrichments for fast throttle movement.  I believe there is a compensation map that some how relates throttle position with MAF.  I'm not sure what it does exactly though.

Here's a graph of the 40% and 68% throttle runs. The purple/brown line is the TPS% and the teal line is the maf voltage. (5.1v is max reading possible.) Notice how at lower manifold pressure the air flow is really high and the afr is really low on the 40% throttle run as compared to the 68% throttle run.

Can you take data from all of the fuel sources?  Base, enrichment, etc...

alfadriver said:

Can you take data from all of the fuel sources?  Base, enrichment, etc...

 

Unfortunately I cannot get the data for individual enrichment, just the overall load used for the fuel table and the resulting final injector pulse width.  I did a boost leak test yesterday and found the BOV and the joints near it are leaking.  I don't know if the leak is big enough to produce these results but the rate of leakage does increase as the pressure goes up so it is probably a factor.