Surging

[stickerbuoy]

Ok, here's a puzzle for you to crack. It's done this since I've had it (180,000-225,000km). Only happens in colder months, and only after it's warmed up. When I first start it idle is fairly high and steady. Once it warms up it starts surging. If I drive for a while, the frequency of the surging will be higher (up to 2 peaks and troughs per second), and if I drive and then park but keep the engine running, the surging will gradually decrease in frequency, eventually to the point where the period (one peak and one trough) is a couple of seconds.

Here's a video of it:

[fixkick]

the ISC is out of control or bad AFR AF/R (fuel mix is way off)
Red 16v :
out of control happens for many reasons.

please post surge range.? this is key, fact.

that surge is the ISC (ECU slave device) surging to hit 800 RPM, what is your Surge range, 500 to1000 , 1000 to 1500 ? or ??????????????
the ECU is programmed to hold 800 hot. and if it cant do that some ECU surge and others just freeze at the limit, (software differences by year, )
if the ISC is not causing the surge, then the AFR (air-fuel-ratio or fuel mix) is way too lean, all gas engine surge like that even carbs (lean).

there are tests we can do to prove that.

the ISC can cause the surge, for these reasons., and more. (in each case the failure puts the ISC out of control (out of authority its called)
1: lean AFR, and RPM way too low,
2: so rich AFR engine has gross misfire. and RPM way too low.
3: EGR sticking open. causing rpm way too low. (way too low means ISC can not get it to the perfect 800 rpm)
4: Intake, induction path has air leaks causing #1, all hose, and air pipes from MAF to intake valves.
5: RPM way too high, ISC cant get it to 800, so surges. , IAC (thermal valve ) stuck open and idle way too high.
6: throttle cables sticking, (up to 3 cables must be set loose, throttle, cruise, and A/T kick down cables.. they must not interfere with the throttle at idle.
7: someone messed with the throttle plate stop screw ,an the butterfly plate is no longer closed, it is must be. its 99.9% closed.
8: someone fiddled the idle duty cycle speed bleed screw on top of the Throttle body, (black cap plastic under it) and its was set way off , so bad
the ISC can not get 800 rpm.
9: the ISC might have its range limited, it still works but it can open all the way or close all the way, so fails. clean it.

TESTS:
im sure if you unplug the ISC it will stop the surge. (electr, connector (has lock lever dont force it , suz, dont sell em)
what is RPM now? hot?
then pinch the ISC air line.
what is RPM now.?
put back the connector
what is RPM now, still pinched? hot.
this test tells me the total air flow , natural, and a cause of wrong.


here are all the air paths,

on a hot engine the IAC is 100% closed.
The throttle valve the same.
so the only air supply is ISC and the bleed, if not?, it will surge, or freeze a the wrong RPM..
all air is metered, if not? , it goes lean, when not in closed loop.

---

im glad its a 16v its Throttle body is simple, unlike the horrid complex 8v TB., that love to leak in very odd ways.

if you screw in the bleed screw 100% and the engine races way over 800 RPM , you do have air leaks, One more test, pass or fail.
even injector base cushions can leak.

mine had a cracked, main huge air intake tube (near #3 in photo), (all aluminum) the back rear alum. base mount was broken, and it sucked in air right there, at the break. took mirror to see it, on car.
as the leaks add up,, it exceeds the authority of the ISC and can hunt.
you can change the Autority rage (called dutycycle) with the bleed.
in fact screwing in the bleed can hide small leaks.
or unscrewing it for idle too low.
the bleed is actually set in the factory for 50% duty, it can increase and decrease RPM equally, by 50% . say 400 RPM to 1200 RPm (crude guess, as there is no spec on this)
if your car was new.and you have a leak of air , more the 400 rpm worth, the ISC might surge, because, it cant attain, at least 850 RPM. if it can't some ECUs surge.
one guy said mine never surges, with on purpose leaks. sure...
that is because your software is different. some ECU do, others don't, ive tried many, some just rail and freeze there.. ECU waiting for ever, hoping RPM drops. to below 850.


rule 1: hot engine.
metered air leaks cause racing (1 example is IAC stuck open or throttle stuck open) if too much, it can surge.
un-metered air leaks causes lean, an surging. ( say a crack in any tube or hose.)

be glad car is not throttle by wire, omg. complex.



unmetered air is air the MAF can't see, the maf is the Air meter. (mass air)

---

i need RPM range first. surge range so i can point more directly.

if car was 1996 and above. we can check for closed loop at idle to see if AFR is right. but no scan tool here.

[stickerbuoy]

RPM range is 1500-1900

[fixkick]

huge errors.

so i bet if you screw the bleed screw in all the way it never locks in at 800 and smooth,
that is sign of
the IAC is stuck open (180F coolant and good hot engine) this valve is 100% thermal active.
or the throttle plate is blocked open some how.
all other air leaks cause too low idle.


what i do is block this hole here.
with putty or tape somehow. (my 97 had this bad. and was stuck 1/2 open)

---

you cant run the engine with the pipe off or the MAF goes to limphome,mode, and no idle controls, etc.
so i had to block it , then put pipe back and test, then remove it later.
if this valve stick open it will scream. that hole is huge. there. lots of air possible.

[fixkick]

i forgot one, dizzy timing set all wrong can wreck idle speed.
and is a this odd chicken and egg deal, the freeze jumper is dead of idle is way too high......on some ECU's.
the timing needs to be near spec, 5btdc or as hood shows.

on a good system. if you advance the dizzy , rpm rises. the ECU uses this magic to trim idle speed fast , then corrects it with the slow ISC, as it catches up, most ECU do this trick on all cars.
so if timing is in ball park.
and bleed screwed in CW. and idle screams (oops got leaks i do)
and i thin pinch the not hot ISC hose ,(air) the idle on a good car drops to like 400 rpm or stalls.
it does that because you kill every drop of air.. its all gone.

the beauty of the 16v, (many) of the TB above, there are only 3 paths for air, and one is 99.9% closed (the TV Throttle Valve butter fly )
the 0.1% is to prevent jam or bore wear.

one guy here, had a racing motor with no air...... (a Kodak moment...... shocking to see)
a very sick , mangled 8v throttle body, your TB cant do that trick thank gosh.

the IAC leaks below TB (missing in photo above)
the TV is not closing
Bleed screw, (now closed)
The EGR port top of TB, will not do this. (it's very tiny)

The ISC on the front of plenum is stuck open. (pinch the hose see what it does)

16v is very easy to find, compared to any 8v or any speed density engine they are all lil horrors... on all cars. with them.


The MAF mass air density system is the best system, it adjusts for any thing. (its only fails with air leaks and goes lean and slower RPM)


there are always exceptions. in any complex system.
here is one , example'
say the system is rich full time, very rich, (failure 1) (pretend fuel pressure is say 66 psi, bad) (top symptom; is no CLosed loop at idle. fix that first)
then you get and air leak,, and it races. air + fuel = race, and it will.
its not supposed to, but does. (physics wins, ECU loses)

but you can starve all air to any 16v engine here easy and prove who works and who don't.


so if EFI is horrid rich
and the intake manifold leaks air it will race , and nothing can stop it. but our test proves that out.... easy.

my guiding light is Greg Banish
bless you man.

---

btw, my gut feel is your fine video sounds like EFI (ECU) caused surge.
if it was lean bog surge it be funny sounding. like it was chocking to death. (hot)
it sounds happy, and full of pep. to me. best i cant tell with cheap speakers...


it's ISC out of control and the ECU madly trying to gain it back...
this is a servo loop and they all can do that. given all ways to fail.

imagine farmers wind mill, water pump (i'ts not a mill but we call it that) it pumps water to the water cistern. for farm house or cows troth.
and the fan blades all point in to the wind and makes max power for the pump.

one day a drunk albatross hits the tail section and breaks off the tail and is now GONE.
the fan just hunts back and forth (surge) never to find the correct wind direction again.
this is servo, too. (100% mechanical and easy to see)

the ISC is the slave of the ECU, and has +-400 rpm authority (shooting from hip spec), if the ISC is not dirty.
if anything exceeds that authority, it will fail and may hunt. (others stick high or low, depending in failure mode)


I used to work on RADAR (huge) servos, they too hunted, sometimes.
Its also bad programming.... (some will say)
on a new Toyota (lawsuits) the new cars go to limphome (slug) at the first hint of idle issues.. rendering car near useless. 15mph car?
The beauty of this car the parts can be all adjusted, and each part can be bought , unlike new cars.$$$$$$$$

[stickerbuoy]

The temperature dropped down again so I was able to do some more testing. I replaced the IAC with another unit that I had, and it still does the exact same thing, surges from 1500-1900 RPM after a 10 mile drive. Seems like -15 or -16 (F) is roughly the magic temperature. If it's much warmer than that, then there's no surging...

[fixkick]

review of facts:

only cold air surge (air temp , not engine temp.) only below 15F or so....
RPM range is 1500-1900 (16v OBD1)
and only after the engine is warmed up fully

the magic temp to me, is obvious. the idle speed is too high for the ISC to gain control ( due to air leaks or duty cycle set wrong)
that 1900 is the key number, in that that may be too high, for even a perfect 50% duty cycle to control.

many of theses cars have bad parts this old. and some are even hacked up. (TPS set wrong, idle stop violated, ISC screw violated ,etc)
finding out which is bad is not easy, but i have a rote path that i follow.

I work the 4 paths is legal air.

hot engine tests:
what is RPM with ISC unplugged?
what is RPM with IAC blocked (no change?> the IAC is GOOD)
those 4 legal paths are: (seen in the drawing above clearly, all other paths are illegal, and i will skip PCV as its tiny.)
1- The TV (throttle valve) and TPS is closed signalling Idle is true.
2: the IAC
3: the ISC
4: Bleed.
what i do is block each one by one to see what happens, ??
Begin testing:
what is rpm with bleed screw all the way in. still way too high? OVER 800 IS BAD.
what is RPM with TV 100% closed, (if it looks closed it is. but it must be. 99.9% closed. (gunk behind it holding it open, or throttle cable set wrong?)
what is RPM with IAC blocked, (no change means its not stuck open, ) this valve is closed 100%# hot.
now we are at the ISC (electric):
we can do 2 ISC tests here, pinch the hose (air hose not the hot water hoses) and see if engine near stalls. like 400 rpm is normal pinched
if that fails that, means you have gross intake leaks. (which can change with air temperatures sure)
if that test passes, super low RPM , then the ISC is open for some reason.
so unplug the electric connector to it, it must now CLOSE 100% (some leak tiny amount , that we can adjust for with the bleed)
it must close, if not, it stuck open. if the pinch test passes. see why, i know that?
I know this because the pinch tests blocked all air supply to the engine, in fact the 1 steps above blocked all air to the engine now and must stall.

this is my tricky but easy way to find the air leak sources.

The hunting is just the ECU cycling the ISC to get to 800 rpm (a goal) and can not get there, so just hunts it , it has no other choice,
it just keeps trying.
btw on new cars it dont hunt, it just slams the system to limphome, or if too bad stall the engine,.
on some kicks/Vitara's the ECU just closes the ISC 100% and prays it will close, (varies by year, engine type and country)

once you prove , say, there is an illegal leak to the plenum (large air box behind the TB , throtte body)
then my pinch page finds the cause.

http://www.fixkick.com/IDLE-AIR/leaks/Hi...-List.html


why 15F?
that is the magic temp that causes the 50 % idle duty cycle to lose control.
The ISC can NOT control idle with vary bad air leaks. it can not.
either kind, measured air or unmeasured air.
it can not.
the best place to have the natural idle is 400 to 1200 rpm. 800 is the center point, at 50% duty cycle.
so if you get leaks say, that cause 1201 rpm the ISC cant correct that, because that is outside its COMMAND authority, range.

lets look at this another way, idle is bad.
so you jump in to the FSM and check idle duty cycle, and find its way off and railed. it's out of control, because of leaks.
back in 1995 the Suzuki shops., tech1 tool shows Duty is railed....oops...
The ISC is not for leaks, its for normal conditions, (hot or cold air, , extra accessory loads on the alternator, power steering stiffer cold,etc)
not for air leaks at all.


I use the nature hates vacuum thing.
The plenum is PULLING 19 inches of HG vacuum. and is 1600 CC strong pull. against a closed TV.
it will find leaks easy, if things go wrong.

if the leaks is from direct atomosphere the engine goes lean (slower) but the hot engine EFI will try to correct lean, but may lose that battle too.
the legal leaks are from the MAF. path. those ALL cause RACING.
the legal leaks are ISC stuck open, IAC stuck open, TV stuck open, and bleed unscrewed (open)
if all my tests fail to find the leak cause
that means the plenum is breached. gaskets seals , hoses, or even cracks.
the process of elimination proves that.


conclusion
there are leaks. but the ECU dutycycle hides them until the air temperature is very very cold.
or might be just duty cycle set wrong.
1: iSC hex screw set wrong, its a factory setting.
2: TV set wrong, same deal.
3: bleed set wrong, in the factory the above are already set, so they do the Duty cycle steps in the FSM. they set it to 50% and 800 happens by
magic. that magic is ECU controls.

The fSm shows using a factory tool to set Duty, and is true. and very hard to do, with wrong tools.
what i do is very crude,
hot engine, 800rpm perfect idle (dont prove its right)
i pinch the ISC, if the idle drops only 100 rpm the duty is wrong. 800 to 700. bad.
if it drops to 400 to 500, its right.

the ISC is bipolar, it must be able to drop 400 or add 400 RPM , if it can do that , all is happy/

[stickerbuoy]

I pinched the hose on the top of the IAC, and the surging stopped. I followed the hose back, and it comes out of the air intake pipe just after the MAF.

My manual says that the ECM calculates the valve position based on the battery voltage, coolant temperature, engine load and engine speed. Hmm...

[fixkick]

yes, all that is true, but temp is 180,, and engine load means, alternator load at idle. and it controls engine speed not the reverse. (at idle hot)
sure as battery voltage drop the duty cycle is increase to correct this error. (all this is transparent)
the ECU uses the ISC (IAC in some books but you must have a name for both, IAC electric and IAC thermal works)
but the ECU uses it to hold RPM at 800 or 1000 for A/C on, or P/s overload.
sure it stopped, but more important what was the RPM pinched, this tell me the total air leaks.

the ISC can not close more that 100% so if there is an air leak causing 200O RPM , it can not get to 800 , it can only close the valve and note that it fails, and tries over and over.

what RPM pinched hot.? 400? if not you have air leaks,
and did the ISC (iac electric) with the connector pulled give you the same RPM , but not pinched? the valve closes unplugged, so SAME RPM. (mine does 400 RPM) .
what RPM?
if you block all paths , there is no more air, and cant run, i saw one guy block all paths and it idled at 600 rpm. he never got that point, gee hows it getting air with no paths for air.
You have air leaks. The hunting is a victim. its not the problem its only a side show, fix the leaks. or calibrated it for the leaks, (the bleed does just that)

there are only 3 more paths normal.
1 down 3 to go. per above.

all i do is block them 1 by 1 to see if any leak.
IAC (thermal) in the Throttle body most not leak hot, or it's bad. if it leaks bad the ISC (electic) will fail totally or at odd moments. (like yours does)
the bleed must not be wide open, close it. see if it stops the hunt.
and make sure the butterfly is closed , TV valve.

if

[stickerbuoy]

This is the part that I pinched the top line to. Do you call that the IAC or the ISC?