IDLE SPEED ADJUSTMENT, HOT  on any Suzuki old    1991-1998 1.6L engine. 

 G16A and B engines. (the DLC varies , 4pin and 6)      Jump to Begin now:

Preamble and theories:

The real title is "Duty Cycle setting". (the ECU controls idle speeds at all times, if the ISC works at all.)

There is no idle speed adjust at all on G16's, you can only center the range of  ISC control, with the DUTY CYCLE adjustments.
Don't be fooled in to thinking the bleed screws idle speed, if it is, then the ISC IS DEAD! (dead means , it is not Regulating idle speeds, now)

The idle switch must be closed at 0volts or this whole page is useless.  TPI and the MPI.

There are vast reasons for the idle speeds not to be 800 (750-850 is spec) and not regulate, the ISC regulates speed near 800, and is hard coded in the ECU to do that full time at hot idle.
(vacuum leaks, ISC stuck, Idle switch stuck open (5v)engine not really hot at 180F or more. IAC stuck open or is not really hot, and more...)
If regulation fails, or the RPM is not 800, or near (50rpm is spec variance) then the ISC has failed, this is an ECU servo loop action. (dogmatic servo loop)
The ECU watches RPM at all times, hot, and  if not at 800 due to loading factors on the engine (alternator loads)  or  driver moves the A/T shifter from park to drive, the ECU sees the RPM drop, then opens the ISC to correct this error, this is a software loop in the ECU, it is a dogmatic loop, endless..... and the net result is idle regulation to 800 at all times HOT (A/C off, and P/S not in overload)
There are only 2 sources of air,hot engine,  the ISC and the bleed screw, so if you have air (vacuum) leaks, you may never get this servo to work at all.
The ISC can only open 100% and close 100% , if this action can not reach 800 RPM the ISC servo CONTROLS FAIL . (means idle speeds  are out of CONTROL of the ECU)
I just found a Duty meter that works, see the sparts list.  $25 tool + RPM and Dwell.

 Failure causes (vast): I can't set the duty cycle. (or idle is out of control HOT)
  1. All fuses good and when the ECU is scanned, code 12 is flashed, 1996+ no DTCs stored.
  2. Engine not running right, running way rich or lean fueling rates, or Failsafe mode, kills idle controls.   Diagnostic mode shows code 12s, and starting in 1996, NO DTC's stored, running.
  3. Spark timing way off, or the cam timing belt slipped.
  4. The TPS idle switch not closed and at 0v. (if at 5v the ECU suspends all idle controls, its not at idle) TBI and  MPI
  5. Engine coolant not at 180F or more
  6. IAC stuck open. its closed 100% on all hot engines. if not , idle speeds "ISC" can never regulate, ever.
  7. Hot coolant water never reaches the IAC and the IAC hangs open. The cooling lines to the IAC thermal valve can be clogged and the valve runs cold and leaks air. (correct water path blockage now)
  8. Gross engine misfiring. for any reasons.
  9. Someone turned the bleed screw all the way in CW or all the way out, trying to cure gross idle speed errors... making matters worse.
  10. The ISC is gummed up, or its internal rubber seat worn out  (leaks) or is sticking as all do from gum up.
  11. The ISC air path is blocked (hoses)
  12.  (the wrong base gasket or hand made, on the throttle body TB?)   the 8v TBI base gasket must block the 5 tiny holes there. or it will fail.
  13. The 8valve TBI body center grasket can crack , suck in and now the bleed leaks air like mad.  and fails for idle speeds way too high.
  14. Some one molested the TV stop screw, (it's a factory setting) If the Throttle valve is not 99.9% closed or 0.0005 (1/2 a thou) at the bore, it's wrong.
  15. Throttle cable not set loose as the FSM shows.  (same with cruise cable or A/T kick down cables sticking and causing TV to not close.
  16. Any vacuum leaks from any source to the induction plenum.  (on 16v MPI this can cause a huge lean condtion and low idle speeds)
  17. Engine running lean and slow.  for  any reasons, or misfiring for any reasons, including stuck open EGR main vavle.
  18. If the engine is running perfectly , now is the time to set, Duty Cycle to 50%
  19. lots more...  under the title engine runs poorly. (ECT unplugged?) near endless, ways for this to mess up.

Common failures: ( idle speed out of control or can't set duty cycle) HOT ENGINE ONLY 180F+.

  • Large Vacuum leaks to the main induction,    hoses all hoses can leak as can all devices on the end of them all of them must not leak air. (PCV a tiny bit, but not hose)
  • Idle switch stuck open 0v good , 5v bad.  TPS not calibrated is the cause,  search in my search box TPS.
  • A/C switch stuck on. 1000 RPM 
  • P/S overload switch stuck on, 1000 RPM  
  • The 1989/90 add head lights stuck on or defrost to above.
  • TV (throttle body throttle valve) stop screw molested.  1991+   it must be 99.9% closed,  (Dash pot retracted where fitted)
  • IAC stuck open or fails to be hotter than 150F (180f is normal)
  • TB (throttle body) screw turn to max or min,  by some person guessing at duty cycle. solution, is to set duty per below.
  • TB /TV throttle cable not set 10mm lose like the FSM demands.  there are cables,  a/t kickdown, cruise and the pedal throttle. all set same way. never tight.
  • 8v TB center body gasket sucked in and now the bleed screw dilling leak air, even with Bleed screw closed 100% CW.
  • The base gasket leaks on TB.
  • The 8V TBI base has 5 tiny holes,  all 5 must be blocked with the correct gasket or all hope is gone.
  • EGR valve body cracked or its base gasket, or EGR stuck open whole or in part.
  • 1996+ EGR sidewinder SST tube damged, or its 2 end gaskets (made of steel) mising or cracked.
I'm sure I missed something, sure,  MPI injector cushions cracked,  or the 3sp A/T vacuum modulator hose fell off or cracked, 
Any leak in the plenum not from ISC and the bleed, is WRONG.
 
Below is 16v MPI engine the Yellow words are don't touch me.
 The bracket has cut angle tab blocking the adjustment screw, on purpose, leave it alone.
It's blocked for a reason. The screw is found on all engines 1991 to 1999 on the throttle body TBI or MPI engines.



The TBI 8V screw is HERE ! for all 1991 to 1995 engines
The 8v screw is painted white and the screw has a plastic cap , warning you, DON'T TOUCH ME.
Materials/tools : needed: ( depends on your tool crib! ) 
  1.  You will not have the very special  &  super RARE factory duty cycle meter.
  2. The MST-2800B  seen on EBAY.  IT WORKS ! I've tested it against my scope and at wide range of car battery voltages, here is a photo of this  nice meter
  3. A modern fast DMM meter , might read 7v at 50% duty but MOST DON'T work, unless it's a very fast modern meter like the Fluke 87v/88V. $$$$$$$
  4. The 1996 + we can use a voltmeter at the DLC pin , duty output. per FSM. chapter 6E3-C2-page 11
  5. The Early cars, show this information, see page, 6E1-78/79 (FIY)
  6. Tachometer, is nice to have but not needed if the dash of car has  one, the 4wd cars DO.   if the RPM is near 800, and regulates you can bet its 800 for sure....
  7. A Scope is always better. (across the ISC , 2 pins) not at the DLC pins ! but costs $50 used or $170 new.
  8. Most simple cheap DMM meters on 20vdc scale fail flat. In all cases , #2 is the best bang for the buck.
  9. Backprobing needles.  or in a pinch leather sewing needles for Walmart for a buck.
Use back probing probes to touch the 2 ISC connnector pins,  on a running engine, do not remove anything.
BEGIN:   DUTY CYCLE calibration.  (the most easy way is here.)

  1. Engine runs great, no vacuum leaks  engine is fully hot 180F +  (engine does not misfire, back fire or bog , on demand)
  2. All accesories off, all elecrical items, head lights, blower fan and a/c off. no defrost.
  3. The idle switch must be closed at 0 vdc  TBI and  MPI (if not stop, dont waste your time below)
  4. Pick a tool,  I'm sure you don't have a scope but if you do , set the bleed so the high and low are .0025 seconds,( 2.5mS)  end job.  (back probling)   t
  5. The scope tip touches the green wire, and blue to scope shield clip.the scope must be ground busted (isolated) or the cars hot wire will be earth grounded. (I'm sure you don't own a differential probe)
  6. You have the MST-2800 , back probe the 2 ISC connector pins (yes do not unconnect any thing) turn the bleed to 50% , end job.  meter is across both ISC pins,  blue is plus.(red test lead)
  7. You have a very fast DMM meter, $$$$ and you can do same as above, but turn bleed to 7vdc, end of job. most meter fail doing this.... most..... but not the $400 Fluke 87v
  8. If no  meters at all , or scope, turn the bleed till engine runs 800rpm and then turn 1/2 turn more,  hope and guessing might work.?
If your efforts work out, the idle speeds will stay at 800, with fanblower and head lights flipped on all at once, or if an automatic trans, holds 800 RPM hot,  when shifted from park to DRIVE or Reverse.
Some later cars are more easy to set, try this.... 1996+

 The newer cars using this FSM. chapter 6E3-C2-page 11, here (96-98' same)

 The early cars used a  custom Suzuki current duty meter,  impossible to find, or buy.


A cold engine can races at 1500 RPM there is no stopping that. (the IAC is physical) unless hacking that is fun.... some do.. (it may race faster in Alaska) COLD.

We have no spec on the IAC except closed at 150F. I have no idea who far it can fully open say at -40F (or C)?
However I did this. at -10F

ISC Tests:   Same basic method on all 1991 and newer.   The 89/90 idles up for Head lights and Rear Defrost too.

And theory is all in the above link.

Best other postings on this subject:

http://bbs.zuwharrie.com/content/topic,31656.msg333371.html#msg333371
Figure 4:  (scope  signal of actual IAC valve.)  It's off a bit,  all you need to do is make the wave equal, for 50% Duty.  2.5mS up, and 2.5 mS down. 5mS period. (200 Hz)
Below the ISC is closed too long, so cut down the bleed, less air on bleed is CW turned. The bleed valve is just a simple cone-screw air valve.  (CCW is more air)
If your idle is at 800,  and the duty is in the ball park, all is well do not over do this..  This setting will not cause AFR to be wrong.


The above is live ISC  signal with my scope,  200 Hertz. Connected directly across the IAC solenoid coil.  (not it is set wrong,,,, but engine is cold , so is normal.)
The scope is ground busted (floating) so it can't upset the ECU driver.
25% closed  because I made an intake leak, on purpose.
Under the duty cycle test above, the signal would  be set to 50/50.    2.5 high and 2.5 mS low.
Turn the bleed screw CW, to lower bleed air and see the Duty rise, to 50% and stop.

Factory way , Pins for DUTY, are here.  (pin 6 , or pin A as the case my be)  (left to right, early 16v,  then 8v DLC, and the right side is 1996-98 (USA) th
The Duty pin (6/A)  is dead until the Diagnostic Request pin is grounded! 
I've never got these duty pins to work, on 89-95 cars,  I use a scope or voltmeter at the ISC pins. directly connected to ISC pins.  OR THIS METER.
The FSM on early cars shows a current mode, duty mete,r to measure duty cycle, we have no such meter..... so skip all the below for duty checks pre 1996
I'm told the 1996 and newer cars in USA the FSM method  that uses the 6 pin DLC pin 6.


8 valve TBP only, idle switch. Is pin C, per this Factor FSM drawing. The harness wire is blue-white. in all cases. 
The switch can be tested for less than 500 ohm, key on, not running, unplugged.
or checking the voltage on the pin key on running and connected (be  sure dash pot retracts or this test fails)  If this switch fails, ISC regulation will be DEAD.
PIN C must be at 0v (never 5v) or the ISC will never ever work.


16 Valve MPI idle.  Pin 2 must be 0v at idle, or the Idle controls will be dead. (ISC dead, and not regulating RPM at 800 hot) The harness wire is blue-white. in all cases.
You can see pin one is BOTTOM.

This is the new $25 meter that I tested, it works perfectly for RPM & Dwell, (4cycle only) and DUTY.  Sold on Ebay, see materals above.seen on EBAY.  IT WORKS

Works best for 1989 to 1995 G16 engines.  Newer cars we can use OBD2 scan tools and do this. (and other ways, stated in the FSM 1996) ( a direct link)
  • Hot engine!  180F or more, is hot.
  • Diagnostic jumper shows code 12s, no DTC errors stored or live.
  • Idle switch closed, 0 volts D.C.
  • Engine not misfiring,  or running poorly. (setting duty cycle is the last step after an engine tuneup)
  • Using 2 back probes , back-probing(verb) the  ISC connector pins (any polarity works here)
  • The meter , is set to DUTY mode, seen in yellow below.
  • Running at idle, turn the bleed screw until, the meter shows 50%, you are done. "easy  peasy, japanesy"?  ( on 89/90 only turn the throttle stop screw)
  • If you can not hit 50%, there are air leaks (aka: vacuum leaks) for sure. (or ECU idle up pins in the wrong state of logic see below for 1 example set)


I'm testing it here, now...... against a very accurate scope.
This is the  most safe  way for ECU, of all.  Unlike pesky scope grounds and the ISC blue wire, issue landing on earth ground horrors.... or lacking a scope differential probe set.
ECU pins wrong list: MPI example only: 1992-95  G16B RUNNING at idle.
A2, Start Pin? in wrong state  , this is the cranking signal to the ECU,  tells the ECU to open the IAC and inject  for slow cranking conditions and low air supply.  12vdc is cranking and 0v is not cranking,  if at 12vdc running, ECU will get confused.
if A2 is 0v cranking the ECU will be confusedand inject wrongly.
A6, on DLC pin, bad at 0v, 12v is good."test switch term" I call it timing freeze. jumper, might be  left in place or wire shorted to body of car.
A17, Diag. mode jumper  left in the connector? 0v wrong. or shorted to car body , wrong. 12v good.
A18 , 0v is ok, , and is the  idle switch is closed.  5v is WRONG.
A22 , tells the ECU that you are in drive.12v now is a lie. 0v is correct. if at 12v the ECU uses a very powerful idle up duty cycle and makes setting duty impossible.
B7 , Throttle angle sensor (TPS) near 1v (if way off may confuse any ECU)
B8, MAF at 1.7 to 2 vdc?  (the ECU can only  detect dead MAF, not weak, so we use a DMM meter and test it directly.
B20, (it's a fast moving 02 signal, and if way off the engine can go rich or lean, a scan tool shows it working or not... and closed loop. ( if engine goes lean, idle can surge out of control, lean can cause idle to fail easy so can gross misfiring)
B21 is IAT, I think this is not important here. what 1-2v or so, depends on air temp.  2v = 100F air temp, a warm engine heats its own air up to about 140F (run long enough) but not in Alaska..
B9 is ECT and is 0.65v = 180F. lower is hotter and ok.  Above 0.83v is a hard failure here, (thermostat bad)
As seen here. MPI (see page 1 and 2 of FSM)
DSO= Digital Storage Oscilloscope.  GMM = Graphing Multimeter.
OTC =  Over the counter.  (after market ,store bought , not Suzuki bought)
CTS means coolant temperature sensor , aka; ECT. new name is Engine Coolant Temperature  (sensor) The are both the same and have 2 pins.
TV =Throttle Valve or butterfly plate.
TB = Throttle body.
Rev 16.     for 1989 /90 coverages  read the FSM for those years. OR  HERE:

Questions,  post them here: http://www.fixkick.com/bbs2/Suzuki-bbs/forumdisplay.php?fid=2