mfi idle on max, still lean on idle?
 

my idle adjust is at full max
first i thought it was the little screw thingy that didn't connect on the inside, but i opened the pump , and the hex nut just won't go any further in the clock wise direction, i can turn anti clock wise, but that's leaning out


lm1 says it's lean on idle
and rich on load

however, if i lean out the main enrichement , the car stops idling


is this dyi fixable? anyone seen this before?
or am i screwed and is my pump overdue for a rebuild?

Thinking laterally for a moment, if your idle is too lean, might you be getting too much air (rather then not enough fuel)? To put it differently, what's the idle speed your at, and have you ensured that the butterflies are fully closed?

Are the butterflies too worn?

What's the clearance between the butterflies and the throttlebodies when the throttles are closed?

my idle speed is a nice 1000 rpm
and the idle is to lean to adjust with the butterflies, if i squeeze it anymore, it runs out of air to idle

i disassembled the TB's, cleaned all the bores, and then benchflowed them with a unisyn and a vacuum cleaner to get proper correlation
there is no play on the butterfly shafts

so i used the cma suggestion , of setting the throttle stops so the butterflies stick, and then turn back a little till they no longer stick


when i installed it in the egine , it wouldn't idle properly
turns out it was not getting enough air ( disconnect fuel pump, open throttle engine idles )

but this was with a main set to very , very rich
as i leaned out the main , the idle went leaner as well
couldn't lean out more or the idle went away

i kept leaning out till my main afr improved
but it's still rich now ,and now my idle is lean


that made me look at the idle enrichement
which seemed stuck, turns out , it's not stuck, it's just at max inside the pump.


i'll run some more tests, but i think i'll have to admit defeat and get the pump overhauled

Page 24 of CMA states:

“Always observe the following sequence:
First part load
then idle speed
Reversing the sequence results in incorrect values.”

Page 32 of CMA states:

“Reason: An incorrect part load CO value
influences the exhaust-gas composition so strongly that it
can be no longer corrected with the idle-speed adjusting screw.”

I would guess the main mixture screw has 5-10 times the
“…so strongly…” effect as the idle mixture screw.


On the diagram below, the main mixture adjustment (Part Load)
#2 is on the end of the rack and is the basic mixture setting.
The idle mixture adjusting screw #15 is in the centrifugal
governor and affects only the idle end of the operational range.
IMAGE: MFIpumpDiagram01.jpg "
http://forums.pelicanparts.com/uploa...1126133809.jpg "
(C) Dr. Ing. h.c. F. Porsche A.G.


Try setting the idle screw lean (CCW) two turns (12 clicks) and
then get it to idle using the main mixture adjustment. Once you
have it where it will idle, change the idle mixture screw and
confirm it has the expected result. This will tell you if this part
of the pump is working.

Be sure to keep track of your adjustment changes so you can
back-track if necessary.

Best,
Grady

wildo! thx Grady

Stijn, Grady is right on.

I think what you might be experiencing is the fact that you have to set the part load mixture pretty rich for it to run right. If you set it to factory specs, it'll have poor driveability and a flat spot around 3000rpm at part throttle/part load.

When I use the 4-gas analyzer, I set them at 8% CO at 2500-3000 rpm free rev. That usually gets it right. MFI gets leaner under load, so you have to compensate.

check the linkage. Make sure the linkage to the MFI pump is not too long and thus fooling the pump into thinking someone is pressing on the gas.

Chris

73 911 E

i did the 114 mm from the manual, made sure of that one
and also checked that the pump is on full stop when the loudpedal is out

i'll run some more tests , hopefully this weekend, if my oil tank is welded up again ( the mounting bolt in the top left corner broke off )

i'll keep y'all update when i get some more data

thanks for the suggestions

little update, finally got my stuff , and did some more testing


Peterfrans came over, and helped out , the ignition timing is now spot on , so is dwell (whiteout on the timing mark... should ave done this long time ago , at least now i can see the mark with the light)

but it still looks like the pump does not put out enough on idle , it's way lean , and i can't lean out the main adjust , or i have no idle left....

so probably , i'll need a pump rebuild this winter

did have a little scary moment , while on a test run
there's a short righthander half a mile before the garage
60 is fast in that turn , and i'm closing in to the corner at 90+ something

when suddenly my throttle sticks at full open in 4th gear
i noticed it when i clutched in... revs went up , but i caught it in time
i reach for my pedal, hoping to pull it back
but it doesn't work , it's stuck

want to turn of the ignition .... right hand side ... duh , i'm not in my 924
meanwhile the corner is getting closer

left hand, ignition off, brakes, and crisis avoided


the draining tube from the air filter housing , had gotten inbetween the throttle linkage of the #4#5#6 throttle body...

well scary stuff


anyway , didn't have to much time today , since i spent some time putting back my fixed oil tank, but i'll give it some more tries tomorrow , and check how far i can close the throttlebodies before it stops idling ,i'm not convinced i'll be able to richen up the idle enough that way , but it's worth a try

I was in the car at that time the throttle got stuck, my hart rate still needs to come down and I needed a clean set of underware:eek: I already thought Stijn was approaching that bend rather enthousiastically before the throttle got stuck, but in stead of braking we were getting faster.

I was looking at the LM-1 before it happened. It is obvious that something is wrong with the pump. We were getting an AFR of 12.6 cruising along at 3500 rpm. On acceleration the AFR went down to 10. At idle it was around 14. The car backfires through the exhaust on lifting the gas pedal.
One thing needs to be said. Stijn removed the cold start enrichment from the pump, this might contribute to this situation. Also the fuel cut off solenoid and associated micro switches have been removed.

But to follow CMA:

1. Air Cleaner Cartridge: ok
2. Compression Loss: engine has been recently rebuilt with new P/C's
3. Spark Plugs: ok
(Spark Plug Connectors)
4. Dwell Angle: ok
5. Ignition Timing: was wrong, now ok, we had an AFR of 20 before, it went to 14 at idle after setting the timing
6. Fuel Pressure and Flow: not checked
7. Injection Nozzles: not checked
8. Injection Timing: ok
9. Correlation: ok
10. Exhaust Emission Test
a. at part-load: ok
b. at idling speed: way to lean

Stijn Vandamme
testing his MFI
with stuck throttle Hehehe


http://forums.pelicanparts.com/uploa...1127584574.jpg

Best,
Grady

Peter,

I'll have to edit the image.

Brst,
Grady

LOL

it needed a small correction :
http://forums.pelicanparts.com/uploa...1127585234.jpg

Grady,

LOL

The wide open eyes should be on the passengers side, the moment it happened the driver dove under the dash to do something to his pedal cluster

like this ?

http://forums.pelicanparts.com/uploa...1127585922.jpg

at least i took care of business :cool:

probably wasn't too elegant though :D

how would a 2,2S pump work on a 2,4S engine?

Stijn,

The backfire when you lift is because you don’t have the over-run
solenoid working. I encourage you to reinstall that. It is a very
simple system with microswitch, RPM transducer, relay and
solenoid. Not only does the MFI run better but you get better
fuel economy and less dilution of the oil with gasoline.

I agree with removing the cold start spray/dribble fire inducing
system so long as it still starts easily cold. Don’t get tempted to
use the cold running thermostat to get it started. All that will do
is dump a LOT of gas in the oil and damage your new cylinders.

One of the best features of the ’69-’71 MFI is the cold start
solenoid on the pump. It forces the main rack full rich. You
can wire a momentary switch to control it if needed. This way
you can set the cold running thermostat to very lean, just
enough to have the engine run. The ’69-’71 MFI pump front
governor cover with two solenoids will fit the ’72-> MFI pump
or you can mod the later cover.

Peter, e-mail me about your 2.2 pump question – gradyclay@hotmail.com.

“I was in the car at that time the throttle got stuck, my hart
rate still needs to come down and I needed a clean set of
underwear. I already thought Stijn was approaching that bend
rather enthusiastically before the throttle got stuck, but in stead
of braking we were getting faster.”

Come on guys - - - I want you to solve MFI issues, not become
victims of a traffic “incident.” PLEASE be careful.

I don’t see anything that changes my mind about my above post.
Don’t assume anything on CMA. Some movie had the
line “Assumptions are the mother of all f***-ups.” This is very
true of CMA. When chasing an elusive problem, I can’t tell you
how many times I have repeated the same tests – all with the
same results. Too often something can change while you are
testing.

Do a search on my extension to CMA. It has several things
in it that weren’t obvious to Porsche when CMA was written
in ’67-’68. I like to call this “geriatric mechanics.”

Please post your LM-1 data and the conditions. That will help.
Are there some known good parts that can be swapped onto
the engine?

To be fair to the artist, here is the original:
http://forums.pelicanparts.com/uploa...1127613145.jpg
(C) 1974 Bob McDonald "McGee"

Best,
Grady

Grady , trust me when i say , i wasn't pushing the enveloppe

i just did acceleration runs on open straights, and with close to zero traffic around, my brakes have been thouroughly tested on previous days, and prior to this incident, i slowed down well before the turns

that throttle got stuck well before this particular turn, which meant i had time to overcome my surprise, and deal with it, just didn't have time to coherantly explain it to Peter.

Stijn,

I had some conversations with Grady about this situation. We should probably fix the missing cold start thermostat first.

Grady suggested to do as follows: make a little adjustment plate that replaces the thermostat. You cannot run with no thermostat in place. The critical issue is the screw must turn off the cold running thermostat while not damaging the pump internals. Screw the adjusting screw in by finger until it stops. Back out the screw 1/8 turn. Tighten the locknut. You can check for the thermostat being fully off by removing the little 2-screw triangular plate on top of the pump above the thermostat. With a jeweler's screwdriver push down on the mechanism below the pivot. The mixture should not change (push = lean). It is desirable to have some slight movement when you push.

OK, now the cold running thermostat can be left in the "off" position (except for cold starting).

i actually had such a replacement thingie when a started off
but with me beeing not so handy with welding and so on , i simply JBwelded an M5 nut to the plate
and then ran a screw in there

obviously ... it fell off not so long ago
i didn't get the idea that while it was on there, it was working
i think the pivot might be "off" because yesterday , lm1 hooked up , engine idling , pushed in where the thermostat should be , but no change whatsoever....

didn't have more time , but was planning the triangular thingie for my next weekend...

for the record, i still haven't figured out what the thermostat does

is it pushing in the pump when hot, or does it push the pivot when cold??
bit confusing with the spring and all, i'm looking at the thermostat now, even tried heating it , but i'm not seeing much movement either way

The thermostat pushes when hot.

It is unfortunately too common to find a MFI 911 running with the
thermostat on for a variety of reasons. Most common is the left
heat exchanger has rust so not enough hot air gets to the
thermostat. Other reasons include the plastic/paper/alu foil
concentric hoses have become restricted internally, a mouse
nest in the hose, etc. The thermostat elements need to be
nice and clean to function properly.

In normal operation you want the thermostat to turn off as soon
as possible – even at the expense of perfect running during
warm-up. A MFI 911 likes rich – it feels great. It just dilutes the
oil with gasoline and washes the lubrication from the rings and
cylinders.

When the thermostat is not functioning properly (in your case
not there at all) then the engine runs VERY rich. There is enough
adjustment range in the main mixture screw to get the engine
leaned enough to run. The difficulty is (other than too rich) that
the relation between the mixture at WOT, part throttle and idle
is now improper. You may get the part throttle close but the
other extremes will be way off and not in the adjustment range.

This diagram is so everyone understands what goes on.
NEVER GET TINKERING INSIDE THE PUMP.
You can turn a rebuildable pump into something
that the Bosch repair facilities will refuse to work
on. Very expensive proposition.

This view if from the right side of the pump looking to the left.
The thermostat rod is in red.
The pivot is in green.
The part the pivot acts on is in blue.
The spring (below the blue) is not as strong as the spring in the
thermostat. As the bi-metal discs expand from the heat, this
spring in the pump is compressed to where the mechanism hits
a stop. At this point the “thermostat function” is off. As the bi-
metal discs continue heating, the spring in the thermostat is
compressed. This prevents the forces from becoming too great.
http://forums.pelicanparts.com/uploa...1127786260.jpg


When you make a mechanical screw to replace the thermostat,
the end of the screw that contacts the pivot should be rounded
and nicely finished. You will find that the screw will need
refinishing periodically because the pivot is much harder steel
than the screw.

You will find the last ½ turn or so of the screw won’t change the
mixture. Always set the screw where it isn’t firmly holding the
spring assembly against the stop.

Remember:
NEVER TINKER INSIDE THE PUMP

Best,
Grady

thanks Grady,

btw, Peterfrans pointed me out to something else
my loud pedal had been bugging me as wel lately
i just thought it was simply my pedal cluster, but i couldn't figure out what it was

but Peter guessed that i did not have full WOT and finally my little brain figured it out.

for some reason , the linkage from my pedal , to the control rack , is to short, and because of that, my pedal is not "high" enough when not pushed in.... so the travel from out to wot is to short... and i never get full wot when i floor it, so that's not helping either

it's strange , because i never adjusted anything during the rebuild , i think the rod between the 915 and the pivot on the left TB is bent slightly... not sure but i will check this next saturday

pretty annoying not to have the car near my house , can't check things during the week , only in weekends, it's so frustrating...

There is another thing to consider:

How is your WB02 sensor attached? If using test pipe, LM1 will indicate lean AFR on idle beacuse of "backwash"-effect of pulsating exhaust gases. With other words: fresh air is drawn inside the exhaust pipe between the puffs and LM1 would indicate leaner condition that it actually is.

It only happends at idle or near idle, using test-pipe. Full-accelleration AFR is usually spot-on even if you use test pipe.

If sensor is welded on the header and you still get lean AFR's on idle, check for minor exhaust leaks... if no leaks, you are too rich.

the sensor is mounted in the tailpipe with a bung

http://forums.pelicanparts.com/uploa...1127820944.jpg

would the backwash pull fresh air that far in the pipe?

i do think it's reading accurate on the lm1, because there is audible popping when it does read low afr

Here is a likely candidate for your accelerator pedal position.
This “Accelerator cushion” 911.423.223.00 is at the rear of the
tunnel, where the throttle linkage comes out of the tunnel and
just prior to the bell-crank linkage at the transmission.
http://forums.pelicanparts.com/uploa...1127830864.jpg

It can come apart internally permitting the linkage to be longer.

Throttle Cable Snapped....on my '68 911L....i think

Another potential problem place can be the plastic bushings in
the accelerator linkage bell-cranks (one behind the pedal, one
alongside the transmission and one between the MFI pump and
#3 intake.) If a pair of these bushings disintegrate, it can cause
this problem.

Is the hinge in the accelerator pedal in good condition?
Is the bell crank behind the accelerator pedal not interfering
with the tunnel sheet metal?


Start getting back together all the parts for the MFI overrun.
That will be your next project after you get the thermostat
issue resolved. You are going to need to reinstall the “Speed
switch,” the “Micro switch” and the “Shut-off solenoid.” Is all
the associated wiring still in place?

Best,
Grady

tomorrow another day of testing, saving this thread to my laptop
and try to get some focussed testing done... not easy , attention span of a chipmunk here...i'll post updates tomorrow

we need updatesSmileWavy

nobody at the garage this morning , turns out he was off to some shop and cell phone not charged up... so went back home... got call later on .. most likely nobody there tomorrow either

it's a solid PITA to not have a garage in your house, or not to have anything within a decent range either

i'll try and get a day off at work this week, but it's real busy at the moment, so most likely it's going to be next weekend :(

Stijn,

Let’s use the time to make a plan.

Can you make the thermostat replacement screw piece this weekend? If so, do you have the little pieces necessary? You will need:
1) A piece of steel about 5 cm X 3 cm X 2 mm.
2) Two M5 nuts (10 mm hex)
3) A M5x30 hex-head bolt (10 mm hex)
4) Ability to drill holes in the steel
5) Ability to braze (preferably) or weld a nut to the steel piece.
6) Ability to grind the end of the bolt into a hemisphere.
7) Some craft paper to make a gasket.

I assume you have the two screws and lock washers that hold the thermostat body to the MFI pump.

You might want to make a couple of spare adjusting bolts. If you forget to tighten the lock nut, the bolt will fall out and disappear under the MFI pump. Ask me how I know.

An easy way to make a template for drilling is to lay it out on a sheet of paper at 10X scale. You can then photocopy reduce it to the proper dimensions. You can also photocopy on craft paper for the gasket.


If you can’t deal with the thermostat replacement screw this weekend, on to Plan B:
Can you collect all the pieces to reinstall the overrun system? What was removed? What is still in place?
You will need the RPM transducer, the connector and wiring in the electric panel, the wiring in the main engine harness, the Micro-Switch & associated hardware and the shut-off solenoid.

If you have all these parts, let’s inventory and bench test them. We will be able to lead you through the process. When you install everything, you want to be sure it works as intended upon installation and not have to spend any time trouble shooting that issue.


If that isn’t possible, on to plan C:
Let’s address the accelerator linkage. Can you safely get under the 911 at home? I don’t mean it teetering on the original jack. I mean safe jack stands and wheel chocks.

The first issue is to look at the accelerator cushion. Push the rubber gummi bellows back. See if there is any appreciable play between the accelerator rod coming out of the tunnel and the linkage piece to the bell-crank on the side of the transmission. (look at the thread I posted above.) The failure mode for this part is the rubber cushion separates from the steel end piece and adds about 5 mm to the travel of the accelerator rod. It is held captive by the tin cover and that limits the failure to only part throttle.

If this is the problem, see if you can order up the part today or first thing Monday.

This problem is usually caused by old age rubber. It can also be because the throttle stop on the floorboard is improperly adjusted or the plywood floorboard has deteriorated and flexes too much. When you mash on the loud pedal, the stop should prevent the linkage at the engine from going to its full-throttle stop. You want it limited at the accelerator pedal. The last 5% of throttle doesn’t mean much in terms of performance.


Stijn, it would be useful to post more pictures and every little detail of this project. There may be other significant issues that need to be addressed. You have a nice, proper rebuild. You can damage it with improper running during this break-in period. We want you to be successful in having a really good running ‘73S – one of the best of the best.

Best,
Grady

Stijn,

If needed I can make you the bolt on piece this weekend. Stuck home with parental duties while the wife is at work. Let me know.

Grady , i frankly can't do a thing if i can't get to the garage

i live in an appartement, there isn't even a closed parking space
my car is located about 30 miles from my house, out in the country...since in R'dam , there's no chance in hell to even find something affordable, safe or practical to store my car in this area, let alone finding a place to work on it...

so any shop time, is purely limited to weekends... and sometimes the shop just ain't open ... bummer

thx for the offer Peter, but it's no use rushing this , i've missed the summer deadline allready , i'll get to it when i get to it... got all winter to prep for springtime...

Peter,

If you can make this part this weekend, please do so. Not only does Stijn need this but one for yourself will help with your MFI. This is a good diagnostic tool. Let your 4-year old help, he is the next generation for Porsche. My son (now 20 in college) could weld at age six with supervision.

Stijn, What are the things that you CAN do this weekend? The winter just makes things more difficult. Prep starts now.

Best,
Grady

EDIT to correct my lame spelling

no car, no tools, no parts... i can hoover my living room and do my dishes... that's not going to improve my mfi though

What is wrong with the standard Porsche setup with the bi-metal disc's and hot-air tube between the heat exchanger and the injection pump. Works really perfect for me. Sometimes standard solutions are better than home-made one's ;) ;)

Another point is that most likely the speed switch is broken because the transistors legs are rusted through. Mine just fall out of the speed switch housing. There are a couple of very useful threads on this BB which can help you out in replacing the transistors with new one's.

Peter

Peter,

I agree; the original thermostat is much preferable. Apparently Stijn doesn’t have a functioning heat exchanger on the left side. He is currently running without a thermostat at all. This left the engine way too rich. He got it to run by leaning the main circuit but both idle and WOT are still off scale. In order to get it in the adjustable range he needs something to turn “off” the cold running enrichment. The screw adjust will accomplish that.



Here are some links for the MFI RPM transducer speed switch.

http://forums.pelicanparts.com/showthread.php?s=&threadid=87400&highlight=MFI+spe ed+switch+RPM+transducer


posted the scanned pictures, here is the link:
http://www.picturetrail.com/gallery...4565&uid=778009
Keith (porschedude)
Unfortunately this link has expired. This was the Bosch wiring diagram.


Early_S_man posted this list a couple of years ago. I haven’t tested the links yet.
“Several people posted multiple threads on one problem, and that unecessarily complicated keeping track of the various threads!
http://forums.pelicanparts.com/show...ght=mfi+warning
http://forums.pelicanparts.com/show...fi+speed+switch
http://forums.pelicanparts.com/show...=rpm+transducer
http://forums.pelicanparts.com/show...&threadid=78685
http://forums.pelicanparts.com/show...mfi+speed+relay
http://forums.pelicanparts.com/show...tch+replacement
Good luck! Warren Hall”

Best,
Grady

here it is ! the long overdue update

in short , things i did today

- balanced my front wheels, needed 25g on the left front tire, the shaking is gone now
- changed the length of my throttle pedal linkage
the pedal felt better, but the drivability was worse
- fixed the thermostate workaround...
no improvement in drivability
- changed my TB linkage with new ones
and noticed that with pedal out , the TB's were not at their stop
so i disconnected all linkage from the control rack
and then replaced with the new ones (old ones were not adjusting well due to corrosion and old age)

continued setting up the linkage, so everything was at the stops
then put my laptop on my targa bar, connected to my LM1
got the synchrometer in there, and started fidling with the settings

at first a pain in the ass, wouldn't idle, so used the shutoff switch, as a new stop (my pedal was still not connected so the linkage would be unaffected by any problems in there)
so i set that shut off switch, screwed it in so the control rack was slightly open... then adjusted the pump (thermostat workaround)and TB's till the car idled again with everything on the stops...

day was over, so went for a drive, to evaluate

what a difference !!!
still a couple of flat spots, but that's definately tunable now
the car has good WOT, and at 6500RPM ... it kicks like a mule, savage power

all things considered, i was just looking at the wrong place
my biggest issue was the linkage, which was ordered some time ago, and now is properly adjustable... getting the thermostat worked around, i had this previously , but it was poorly made, so it fell off... then the pedal linkage...
so multiple problem troubleshoot... probably typical for MFI , maybe you can comment on that Grady? is that typical?

anyway , tomorrow another day , looking forward to get some finer adjustments made...

one setback though ... stupid mistake on my behalf
went for mornign drive , still foggy , with the targa top on
but the latches weren't locked, right when i noticed, the roof went airborne.... slightly dinged, but still attaches fine... i think it's the cars way of telling me it likes beeing topless :D what a girl !!

picture of my thermostat workaround

http://forums.pelicanparts.com/uploa...1128784933.jpg

the front plate fabrication didn't work out , didn't have the right gear to hard solder or braze a nut on the plate

so went the other way , the top plate was thick enough for a hole with thread... and it was very easy to control the linkage with a screw directly on the adjuster ( instead of using the built in lever thingie)

drilled a hole, pulled thread on it, in with the M5 screw
done...

it's slightly less accessible than the front plate, but i don't have dokworkers hands, so it works for me...

did some more tuning today , get's better everytime
pedal is a lot better, but i'm sure the bushings and what not can use some refreshing, the stuff you mentioned Grady

at the moment, no major flat spots, zero hesitation. but still a tad on the rich side and a lumpy idle, but it's just a matter of refining things a little bit more
i can notice that at 6800-7000 it suddenly finds a great deal more power( the sorts that makes your face deform) and it keeps it up till max rpm


if you wanna see/hear how it behaves right now :

http://members.chello.nl/s.vandamme1/takeoff.mpg

looks a tad slow ,but that's the angle the camera has to the road, and the poor framerate, that was actually a reasonable launch, albeit not flatout bouncing of the revlimiter.. it wasn't fully warmed up yet...

so next time i'll be working on the loud pedal linkage
and i'll try to lean it out a tad more, while at the same time improving the idle....

Stijn,

I hope you are keeping track of all these “little” adjustments. Yes it is common to have multiple problems – many self induced.

Good to see some progress.

Best,
Grady

got the shutoff solenoid back on today , works and all that
but the engine is still a bit rough and lean down low, so the shutoff kills
it

i'll redo my procedure once more on saturday , every time i go through it , i end up with improvement


i noticed that with the main gas pedal linkage attached, i'm having a heard time getting a steady rpm , it flutters a bit , which makes it impossible to measure airflow over the individual stacks

so i disconnect that one
use the shuttoff switch stop screw, to keep the revs up and steady
then i measure
close the stops a bit
synchronise them
use the lm1 to see if my mixture richens up
as it richens up i can unscrew the shutoff stop a bit, and my revs drop
then keep doing that untill the engine idles with everything on the stops(pump and TB's) and the shutoff stop is no longer used

once that's done, i can lean out the pump a bit more
time for a drive and evaluation

if necessary, repeat the above steps again


it's the only way to do it , because if i do it to fast, eg , close all the TB's and then run the engine
i end up without idle, or the change is to big and i get weird symptoms
this way i can keep an idle , and have the ability to measure airflow , so i can synchronise the TB's

but so far it works fine, done it 2 times now, every time i get major improvement
WOT is very good now, i think one more time will get me a decent idle and midrange AFR