DaveFJ
|
| posted on 1/8/16 at 08:42 AM |
|
|
some tuning guidance needed
So... after a long slog and a lot of work, she is road legal once more! (been off the road for 3 years.....)
After all the changes I now find its not running quite as sweetly as i would like...
at lower revs and idle I get popping from the exhaust, also the occasional fart back through the throttle bodies - but it has always done that...
checked and cleaned with wideband O2
fixed small exhaust leak
several runs now with tunerstudio in autotune mode
I tried smoothing the ve table and allowing tunerstudio its head, the result is a very lumpy map with a large spike at about 1/4 throttle and 2500
rpm
weird thing is it runs nicely at the spike and above where the map drops way back down. but at lower throttle settings i get quite a lot of exhaust
popping (which i assume is unburnt fuel?) tunerstudio has set something like 1/4 throttle @ 2500 rpm = 108 and 0 throttle @ 2500rom = 5
the details:
Pinto 2.1
MSnS v1 on v3 mainboard ?ay29 code? (recently repaired and checked by Bailey performance)
Mondeo 2.0 coil pack
Suzuki GSXR 1000 TBs with standard injectors
Innovate LC1 wideband
Using Alpha N
New cosworth HP pump delivering a steady 3 bar
Have uploaded the current msq and a datalog from yesterday to my google docs site here:
Google Docs Download Site
Any help and advice would be most welcome!
But no... Unfortuantely I just cant afford a rolling road session as I have spent so much getting her back on the road 
Dave
"In Support of Help the Heroes" - Always
|
|
|
|
|
gremlin1234
|
| posted on 1/8/16 at 09:01 AM |
|
|
best thing for it is a long run, get the engine nice and warm, and blow the cobwebs out
ps nice fresh fuel is always a good idea too
|
|
|
DaveFJ
|
| posted on 1/8/16 at 11:02 AM |
|
|
Had a good run out yesterday for over an hour - on fresh fuel... cobwebs are definitely pretty resilient!
What I cant understand is why my ve map looks like the Himilayas!
[img] [/img]
Dave
"In Support of Help the Heroes" - Always
|
|
|
coozer
|
| posted on 1/8/16 at 03:32 PM |
|
|
How far are you from Baileyperformance? Take it too a good tuner and get to sorted....
1972 V8 Jago
1980 Z750
|
|
|
DaveFJ
|
| posted on 1/8/16 at 04:18 PM |
|
|
Unfortunately the wrong side of the country and also blown all my budget on recent work...
Dave
"In Support of Help the Heroes" - Always
|
|
|
rusty nuts
|
| posted on 1/8/16 at 06:05 PM |
|
|
Have you balanced the TBs? I had all sorts of problems shortly after I converted mine to injection, turned out it was the throttle position sensor
even though it was new.Might be worth testing?
|
|
|
SPYDER
|
| posted on 1/8/16 at 08:20 PM |
|
|
I've just loaded your msq and log into megalogviewer.
Your VE table load axis runs from 64 to 213.
Your AFR table runs from 15 to 106.
This seems to set the top half of your VE table to 12.7 AFR target.
The 108 figure is at 1500 revs. It receives very few hits. I would set it to 64 and the bin below it to 62.
And the 75 to 52.
I assume your throttle ADC goes from 64 to 213.
IMHO the next line up at 74 is too large a gap. I would have it at 66 and keep the next few lines up closer spaced.
I suggest 64 66 70 75 80 90 100 110 120 150 180 210 for the VE load axis.
64 66 70 90 120 150 180 210 for the AFR load axis.
Go for another log run but don't have the auto update running.
I think your timing table is a bit conservative. I'm no Pinto expert but a little research suggests more like 34 degrees at WOT above 3000 revs
or so.
Be sure to save your current msq before making any changes.
|
|
|
dave_424
|
| posted on 1/8/16 at 08:30 PM |
|
|
I believe that TunerStudio has to be configured correctly in order to suggest accurate adjustments. "Garbage in, Garbage out"
I would shy away from self-tuning tools and either try to make adjustments yourself or get a professional to take a look at it
|
|
|
DaveFJ
|
| posted on 1/8/16 at 08:34 PM |
|
|
Cheers... I will give this a go tomorrow 
Tried to check the balance of the itbs but my gauge appears to have lost its ball ..
Was concerned about the spark table but unsure how to fine tunexample that... if I remember correctly I was sent that table a long time ago by another
pinto owner....
[Edited on 1-8-16 by DaveFJ]
Dave
"In Support of Help the Heroes" - Always
|
|
|
coyoteboy
|
| posted on 2/8/16 at 07:15 AM |
|
|
Autotune works well if you know how to tame it. Your Himalayan peak seems to be the default autotune response to tip-in leaning.
|
|
|
DaveFJ
|
| posted on 2/8/16 at 09:04 AM |
|
|
Thing is.. I thought my map wasnt far off... hence I thought auto tune would help guess I was wrong!
Dave
"In Support of Help the Heroes" - Always
|
|
|
SPYDER
|
| posted on 2/8/16 at 09:15 AM |
|
|
Hi Dave. Rather than just "going for a drive" why not apply a little strategy? For the VEAL feature to work well it needs a number of
"hits" on each VE bin.
I start off by editing the dashboard to show a large, digital type throttle ADC gauge so it can be seen easily whilst driving. Then write down the VE
table load axis values on a bit of paper to stick to the dash. 64, 66, 70....
Then whist on the drive I set out to hold the throttle constant at a given value, say 75, in third gear and let the revs rise across the VE table. I
might even left foot brake a bit to hold increase the number of hits on a critical bin.
Start at low revs and get the revs up way beyond where you would during a normal drive. Thats the whole idea, to populate the whole table, not just
the areas used during normal driving.
Lower throttle openings might require second gear to get the revs up the table. And left foot braking to increase the time spent on each rev bin.
Repeat at the next throttle number and so on. At higher throttle openings a long hill comes in useful. And your left foot!
I see you have lambda enabled. I got better results with lambda switched OFF during testing. I know they say to leave it on.
After a few runs at each throttle position you can check your progress using the STATUS function. It shows where you have been on the table and for
how long.
I then view the changes it wishes to make before accepting them. Or not.
I always SAVE the current tune before any testing.
In OPEN LOOP mode you have set lambda control up to throttle 192.
I think this is too high. I would have around 120 once you have a reasonable VE table. MS1 lambda control is notoriously slow and "hunts"
around a lot.
Not what you want at high throttle openings.
Similarly you give it 15% authority. I would suggest reducing this once you have finished tuning.
Your TPS THRESHOLD for accel enrichment is at 0.586
Try dropping it a notch. It could help the "tip-in" issue. Do some deliberate "tipping in" then look on the log for the TPS
dot value you are hitting.
(Mine is at 0.196 but I do have a very clean TPS signal.)
The above strategy worked for me. AFRs are good and the car drives well under all conditions. And it goes like stink!
Coincidentally, I run MS1 V3 and have GSXR throttle bodies although on a Toyota 2 litre 3SGE engine. I run the High Res code. It gives much finer
control on pulse widths, particularly useful at idle.
My spark table is in my photo archive. Take a look at it. It has more advance in the cruise areas than yours and "low 30's" at full
load.
Do you have a vacuum line connected to your fuel pressure regulator, or not?
I fitted a larger throttle quadrant to my itb's and offset it to give a "rising rate" response. Much better control. Pic in
archive.
|
|
|
DaveFJ
|
| posted on 2/8/16 at 11:43 AM |
|
|
Some great points made.. things to try
Now to show my true ignorance!... what does 'tipping in' mean?..
Dave
"In Support of Help the Heroes" - Always
|
|
|
coyoteboy
|
| posted on 2/8/16 at 12:09 PM |
|
|
tip-in is when you get on the throttle after being off it or at a stable throttle position for a while. Like when you hit the throttle from idle, or
when you're cruising and go to accelerate. tip-out is the opposite (from stable to close throttle). These transients tend to generate huge
fluctuations in AFR and can cause stumble or misfire, but likewise shouldn't be compensated using the main table. This is what accel enrichment
is there to compensate.
You should turn off all accel enrichment during autotuning.
You should only tune "steady state" using aututune - i.e. sit at a load/RPM and wait for it to level off it's changes, then move to
another load site.
Then when you're happy you have all your sites mapped, turn AE back on and correct that so you don't get massive AFR variation when you
press the throttle or (less important) release it.
If not, when you tip-in with poorly defined AE and fast autotune updates, it sees the lean and tries to accomodate that in the main table, this then
overly enriches that area (so it's awful in steady state) and means it's variable with RPM and load rates. Bad times.
I spent the best part of £300 in fuel trying to adequately road-tune a turbocharged engine, including learning it's quirks and the autotune
software quirks. It worked a treat but I might as well have taken it to a dyno lol. Good for learning.
[Edited on 2/8/16 by coyoteboy]
|
|
|
DaveFJ
|
| posted on 2/8/16 at 12:36 PM |
|
|
Thanks... that's really helpful
Will try to spend some time this evening looking at these settings.... if it ever stops raining!
Dave
"In Support of Help the Heroes" - Always
|
|
|
DaveFJ
|
| posted on 3/8/16 at 09:09 AM |
|
|
Reading through Spyders post.... would it be worth my while upgrading to the v11 High res code?
Dave
"In Support of Help the Heroes" - Always
|
|
|
SPYDER
|
| posted on 3/8/16 at 10:23 AM |
|
|
Hi Dave. When I fitted the throttle bodies I had a few issues with Tunerstudio not functioning properly so I got in touch with Brian Holzbach at EFI
Analytics.
As part of the solution he recommended that I upgrade to HR11d firmware. Here is part of his email...
" The hr_10 code could make things easier for you but because
> you would be switching firmware families you wouldn’t be able
> to load your settings from your other tune you would have to look
> through all the settings and set them manually.
>
> As far as loading the firmware goes its not a very difficult process
> the biggest pain is if you have custom values set for your coolant
> temp sensor or intake air temp sensor you need to use easy therm
> to reset them.
>
> Here is a link to some instructions from symTechLabs that go over how to
> update your firmware you can take a look and decide if that something you
> want to take on.
> http://kb.symtechlabs.com/megasquirt:flash#ms1_extra
>
> Here is a link you can use to download the hr code if you want to give it
> a shot.
> http://www.msextra.com/doc/general/downloads.html "
I found it straightforward to do. I specifically had problems with low pulsewidths at idle. The HR code has about three times the resolution of
026y.
You may not benefit from it to the same extent that I did. The good news is that your VE table should be OK if you choose to upgrade sometime in the
future.
Brian went on to say that even the HR code is yonks old and I should upgrade to MS2 or MS3.
I plan to go to MS2 in the winter.
It might be worth you contacting Brian regarding upgrading. He asked me what firmware I was running and made a recommendation based on that.
I'd be tempted, however, to get the tables sorted first.
Geoff
|
|
|
DaveFJ
|
| posted on 3/8/16 at 10:55 AM |
|
|
thanks, kind of confirms what i thought..
I recently re-calibrated my sensors so i have all the relevant data to sort that out.
But as you suggest - I think i will wait a little and see how i get on with the suggestions above before going for an update.
Would love to upgrade to Ms2/3 but having blown my budget on getting the car back on the road, i just cant afford it right now
Dave
"In Support of Help the Heroes" - Always
|
|
|
DaveFJ
|
| posted on 4/8/16 at 03:58 PM |
|
|
So.... I made a few changes to my setup last night.. have added the new msq file to the archive at
https://drive.google.com/drive/folders/0B03yZv_mWDgSb1VZVl9vQnpnRGM
took a very short drive doing just a datalog and have uploaded that as well.
I did not switch off accell enrichment last night - because i couldnt work out how at the time! (I know now )
but i did switch off the fuel cut on decel.
I corrected the TPS settings on the AFR table and implemented the suggested changes to VE table. also took a good look at the spark table and tried
to come up with a table based on suggestions here and on multiple different spark tables that i found others had posted for a Pinto. Not sure I have
got it right - in fact pretty sure it needs updating.
warmup is now awful, but i can worry about that once i get the main stuff sorted i guess.
One point i note is that at idle the AFR gauge goes off the scale to lean. Looks more like the o2 sensor just doesnt like whats happening as it snaps
round.
a little throttle and it dances back up again. I realise my VE table is very lean at zero throttle so will try adding a bit extra there....
was driveable and maybe a little happier at a steady 30mph... but definitely some way to go before it is happy.
Dave
"In Support of Help the Heroes" - Always
|
|
|
coyoteboy
|
| posted on 8/8/16 at 05:24 PM |
|
|
Not checked your log but...
Idle AFRs are heavily linked to idle spark advance - keeping fuel the same you can swing 2-3:1 just by shifting spark angle 5 degrees, as it leans out
it increases RPM and jumps to a richer point and hunts about. I preferred to leave a fixed idle spark to remove this feedback loop. Also, look out for
sensor heatsoak which doesn't help with this.
|
|
|
.jpg)
