Mr.Hawver's MS3X Tuning Thread

Mr. Hawver

Member
Feb 27, 2006
32
1
8
Hey there,

I have a MS3X project that is running decently well, but I'm hoping to get a little assistance to smooth things out and optimize the tune. My biggest concerns are the lean start (pegging out my WBO2s, despite increasing the After Start Enrichment significantly) and the bucking from idle to ~1600 RPMs. I also feel like my timing table is a little soft, but read the Trickflow heads are relatively efficient and prefer 30-34 degrees advance.

Engine:
  • 302 with stock bottom end
  • Trickflow Twisted Wedge heads
  • Trickflow Stage 1 cam
  • Edelbrock Performer intake
  • 42 lb/hr Bosch/Ford green top injectors
  • LS2 truck coils (D585)
Electronics
  • MS3X
  • Sequential fuel and spark
  • Dual 14point7 Spartan 2 widebands
  • Slot MAF in a 3D printed ABS tube (with velocity stack and grid to promote laminar flow)
Any insight would be greatly appreciated!
 

Attachments

  • 2019-07-23_08.45.00.msq
    275.8 KB · Views: 126
  • 2019-07-23_07.48.48.zip
    1.5 MB · Views: 94
  • Sponsors (?)


I will attribute the majority of your startup problems to the VE table. being as rough as it is, when you look at the datalog notice the VE1 value and the PW. they are swinging by significant amounts until the car settles into a steady idle.

Your ignition table looks like it needs alot of help, the values should be linear like for the most part from top to bottom

This is more like something i would leave a customer with.. Lets see how the changes do with a startup.

I noticed that your MAF input is not very clean, it either has noise on it or the maf is experiencing turbulence such as fan wash. I reduced the lag factor is introduce some smoothing.
 

Attachments

  • Hawver REV.1.0.2.msq
    272.7 KB · Views: 120
The VE table was quite rough since it was the output of VEAL without any smoothing.

Here's a datalog from my commute this morning with the tune you provided, Steve. Overall, it ran better and the idle was more stable. It did want to die on me when I first started driving it after I let it warm up to 160, but I think adding PWM% @ temperature in the initial value table should correct it. The off-idle bucking is reduced, but more pronounced at 1500-2000 RPM.

Last night I made a cardboard (ghetto) barrier between the cone filter and radiator to see if the MAF noise was caused by wash off the fan. Comparing the datalog from last night prior to the box (also with your tune/adjusted sensor smoothing) and this morning, there doesn't seem to be any discernible difference. The MAF uses the same sensor ground plane as the IAT in the MAF, TPS, and MAP, but the signal fluctuation doesn't appear in any of those inputs.


Rick Hawver?

Afraid not, Fast. There aren't too many Hawvers out there, so I'm sure he's some relation of mine.
 

Attachments

  • 2019-07-24_07.42.41.zip
    1.4 MB · Views: 111
  • Like
Reactions: 1 user
This is a fuel only revision, it will clean up further with a switch to MAF based AFR and IGN tables.... I will need more log information to properly build those tables. How long have you been tuning on the car?
 

Attachments

  • Hawver REV.1.0.3.msq
    272.4 KB · Views: 123
Thanks! I'll load that tune up tonight before I leave work to get a log. How much data would you like to see realistically? I know more is always better. Across the whole RPM and load map too?

I've been tuning off and on since mid-May. I'm on my second MAF tube design after the first one printed from PETG decided it didn't like underhood temperatures and took on more of an elliptical cross-section rather than circular. So with that, I had to go through a couple iterations to develop the MAF transfer curves.

It's been a long process that started has an H/C/I install and battery relocation, followed by the MS3, custom harness, and a bunch of other small projects.
 
Hey Steve, here are a few datalogs using the latest 1.0.3 MSQ.

The log from this morning has a few WOT pulls in various gears. After warm-up it seemed to struggle a little while modulating the throttle to pull out of my driveway and proceeded to stall a few times without putting up much of a fight to stay running. Had I been a little faster on the throttle I could have caught and saved it.

The shortest log is after a hot start from when I ran an errand. It didn't particularly care for the slightly heat soaked IAT temps, but it settled back down after I started driving and got some air across the element.
 

Attachments

  • 2019.07.24-25 Datalogs.zip
    4.3 MB · Views: 104
No wonder things were never the same with each datalog as far as AFR... I kept wondering why I was fighting myself. The EAE has to be the LAST thing that gets tuned, it should not be on during the tuning process. I have disabled it and changed all the AFR and IGN parameters to follow the MAF.
 

Attachments

  • Hawver REV.1.0.6.msq
    272.7 KB · Views: 96
EAE tuned correctly, or did you mean just the VE table tuned correctly, or something else in transient fueling?

I ask, because I've done a band-aid tune that's too rich in the VE table and I used closed loop to pull fuel back out. Otherwise, I just can't keep it from going lean on modest acceleration. My new theory is maybe I don't have the injector-voltage slope right, but that seems unlikely.
 
Last edited:
EAE tuned correctly, or did you mean just the VE table tuned correctly, or something else in transient fueling?

I ask, because I've done a band-aid tune that's too rich in the VE table and I used closed loop to pull fuel back out. Otherwise, I just can't keep it from going lean on modest acceleration. My new theory is maybe I don't have the injector-voltage slope right, but that seems unlikely.
The tune has to be dialed in correctly before EAE can be turned on. I like to say within 3% is pretty darn close.

If you are going lean on modest acceleration try using the MLVHD for BS3 this will allow me to look at the logs and build a histogram using filtered data to see what the enrichment's are doing during the transient... [ acceleration is a transient ]
 
I assume MLVHD is a piece of software.

I'm tracking on the transient stuff. I've played extensively with it to try and filter it out. I know I'm just doing something wrong, because I don't have the issue with a BS3 on my other car. I'll follow up with you, later. It'll be a bit before I get back to tuning. Thanks for the offer, and if it's still on the table later, I'll take you up on it. Enough thread hijacking from me.

BTW, I didn't call you, last weekend, because my buddy's customer decided on a tuner, and we support the idea so that the liability for the tune is elsewhere.
 
  • Like
Reactions: 1 user
Here are some logs from the 1.0.6 MSQ. I look it pretty easy on the two drives last night, but got into it this morning with some near- or WOT pulls.

I noticed last night while cruising that my driver's side wideband was reading leaner than the passenger side. At idle they seemed more in sync. Think there could be a potential exhaust leak behind the sensor in the ball/socket joint throwing off the readings? It seems for a leak it would be opposite, with air being pulled in at low back pressure during idle causing an artificial lean reading.
 

Attachments

  • 1.0.6 MSQ Logs.zip
    4.4 MB · Views: 108