evap code PO0443

[dynabill]

Hey all after a little over 500 miles, on new head, I got a code po443 i checked the pages and was reading here http://www.fixkick.com/sensors/EVAP/EVAP96.pdf i did the easiest check which was to check the ohms on the evap canister purge valve, (I am thinking this is the one with two hoses that come off the top of it.)and ohms are supposed to be 28-36 ohms, that passed with 35.3 ohms. i then checked the evap canister purge solenoid, resistance is to be between 25- 30 ohms. (I am thinking this is the one with one hose comeing off of the top and one off the bottom of it) I failed this test with 33.5 ohms. Should I run through the rest of the tests or is this a dead give away that it is my problem. Bill [/i]

[fixkick]

as easy as loose gas cap.!
33 is ok, keep in mine, if it cycles perfectly with 12v, its ok, 3 ohms will not make much difference. s
shorted coil , bad, open coil bad, about 30 ohms, and works with power applied and not,
lets review 443 in the fsm
"incorrect purge flow". the lt-green/yellow wired, solenoid. ( it says leaking vent hose, and that means gas cap too.,)
http://www.acksfaq.com/96-FSM-v2-6E3-A-part4-pdf.htm
page 132
a tech, would check the cap first, if was loose, bingo.
if cap does not leak. then do a take pressure leak down tests, if that fails fix that first. vast places to leak.
then do the hard tests.
the diag is done only at idle. never driving, only parked. or at idle hot.
and can take 20 mins to clear after a reset, or never on and empty tank 15% min.
page 133 is the test set.
no kidding evap is way complex.
3 chapters in the book here are the next 2. EVAP1 and 2
http://www.acksfaq.com/96-FSM-v2-6e3-c3-a-pdf.htm
http://www.acksfaq.com/96-FSM-v2-6e3-c3-b-pdf.htm

im not sure if you have EVAP 1 or 2 system , some 96 have only EVAP 1 and is vastly more simple, just count solenoids. 1 sol, is early..

im hoping its EVAP 1
the 2nd chapter, page c3a page 1, tells who the ECU finds the failure, and is KEY. see text "

the ECU can also check..

.".


1: hot idle. about 19inches plenum vacuum at the ready. beiind that VSV (closed) then..
2: the ecu cycles the purge valve
3: and then it looks at the ISC duty cycle and at the fuel injection times.
when they open the valve , air flows from the EVAP tank up the carbon chamber and to this point and from the tank past the diaphram regulator back aft.
if the iSC duty don't close down (less idle) and the injections dont rise. then it fails. (this is what the ECU does to do the actual 443 test)

a hand vacuum pump can test the regulator back aft via the front canister fitting and you can test the canister at its fitting. for leak down rate.
it must leak. if not its blocked closed some how.
and you can test the VSV 100 % just by applying 12v. or let the ecu open it. and see it suck air with the hose off on output side of VSV.

when the Purge VSV opens, extra air (and fumes) is allowed in to the plenum ,this causes a lean condition (Unmetered MAF air), the ecu sees that and increases the injections, RPM rise over 800 the ECU then sees that and drops the ISC more closed. and 800 rpm happens again.
when the VSV close ,the reverse.
that is how the EVAP 1 works. very simple,

hope that helps. EVAP2, if you have that, its a near endless story. but do tell !

keep in mind the gas cap has a vacuum breaker so of the diaphram valve at the tank line sticks open , the vacuum can suck too much.
that is why they have you check the freeze frame. to see if its
1: too much flow or not enough.

[dynabill]

hey Jerry thanks for the quick reply. I started testing the evap canister, and I cant blow air thru either pipe. the tank pipe is the one coming from the passenger side, and hooked to what looks like a gas line, which they are calling the tank line, gas tank I finally figure out, haha. So maybe this is the culprit? it is old and beat up looking evap canister. rock wants $143 for this gem, guess I will take a look at the junk yards, lol. Also, My first thought was the gas gap, but when I unscrewed it the first night it threw the code, I had good pressure escaping from it. Is there a test for the gas cap?
bill
p.s. I just ordered the Autel MaxiScan MS509 from amazon, be 5-8 days before I get it tho.

[fixkick]

hey Jerry thanks for the quick reply. I started testing the evap canister, and I cant blow air thru either pipe. the tank pipe is the one coming from the passenger side, and hooked to what looks like a gas line, which they are calling the tank line, gas tank I finally figure out, haha. So maybe this is the culprit? it is old and beat up looking evap canister. rock wants $143 for this gem, guess I will take a look at the junk yards, lol. Also, My first thought was the gas gap, but when I unscrewed it the first night it threw the code, I had good pressure escaping from it. Is there a test for the gas cap?
bill
p.s. I just ordered the Autel MaxiScan MS509 from amazon, be 5-8 days before I get it tho.

the gas cap, has a vacuum breaker. if its good and not allowing air to go from inside cap to out side its good.
if you get the Whoosh sound its holding tank fumes and pressure. that is GOOD cap.
the vacuum breaker stops a catastrophic vacuum collapse of the gas tank.
At the smog station they test the cap , then fit a pressure tester to the gas tank, and see if it holds pressure, (im not sure but guess at 2 psi?)
if not they fail you, not even turning on the key yet, (EVAP 2 finds this error too, and is why EVAP 2 exists, and is used till now 2013)
Evap2 system has 3 solenoid valves, see B chapter last post.
the FSM is very weak on EVAP 2 testing. very much. but the do have you blow air (use the vacuum tool) from VSV out to VSV IN (purge)
keyoff , no air must flow thru the VSV;
i wish i know which system you have, 1 or 2 evap.?
1 solenoid or 3.?
there are lots of tests, on EVAP2 , way more. lots of things can fail.
1: the tank, all hose, all parts. related, leaks, or bad vacuum lines, or bad VSVs (up to 3) bad reg, bad gas cap. or ECU controls dead or stuck.

let me know how you do.

---

to correct the FSM.
1: Do not blow or suck with mouth or lips, any fuel parts (benzene is a known carcinogen) Keep in mind the EVAP canister is Benzene trap, it saves up this nasty solvent, one of the worst)
2: but do use a mityvac or HF clone hand vacuum tool , it can do wonders.

the tool can be used to test, all parts,
1: vsv 100%
2: all hoses, , plug one end ,suck other.
3: diaphram leak down tests.
4: you can test that the canister is not clogged up.
5: it can test the reg./check valve near tank. this vent valve, only sucks from front to rear. do not suck from rear of this valve, because that is closed at all times, in that direction(check valve action)
but you can in fact suck the front hose going to the REG. and IT must start up to open at a low secret vacuum (it's just a simple spring valve in the forward direction) (there are 2 tests for this valve, check action and not stuck open)
i dont think you can test the tank air tightness with such a tool, that takes a pressure test, using shop air.
I use a modified, gas cap (added air nipple) and a low pressure paint spray reg. set to 1 psi. (shop air system) then pump my gas tank up; 1 psi to see if it holds pressure, if not , i find the leak, (usually only EVAP2 systems need this harder test, or tank fumes, make me nuts)
to find main fuel tank leaks, pinch off or plug all hoses to tane, or use soapy water spray on said hose to see bubbles. (or with hose to EAR, play it around parts to find a hiss sound)
block the reg , if that stops the leak the REG is stuck open. Reg means , Tank pressure control valve.
on EVAP2 it is a horrible tricky device. testing a 3 hose REG ,i'd not try, id replace it when all tests pass above. But finding one stuck open is easy.


sorry for long post but vacuum and air leaks are the most complex of all problems, to cure.

IF Evap 1, ?
i try first to see if the ECU is finding NO AIR flow, or too much. knowing what it finds, is the best clue to the solution. in every case. (but can be hard to discover this)
These old cars have piss poor docs on ECU actions. for most OBD monitors newer cars they flat out tell you , all tests done. and precisely how.
but we can sort it, for sure.

[dynabill]

i wish i know which system you have, 1 or 2 evap.? yes me too, lol. I know my way around these pretty good, but this is new territory.I checked b chapter and I dont have three hoses coming off of the evap canister. this is the only difference i could see in them. although I do have a nipple on the bottom of the evap canister that has no hose for it to connect. I am pretty sure, my 95 had a hose that it connected to.

1 solenoid or 3.?

there are lots of tests, on EVAP2 , way more. lots of things can fail.
1: the tank, all hose, all parts. related, leaks, or bad vacuum lines, or bad VSVs (up to 3) bad reg, bad gas cap. or ECU controls dead or stuck.

let me know how you do.

---

to correct the FSM.
1: Do not blow or suck with mouth or lips, any fuel parts (benzene is a known carcinogen) Keep in mind the EVAP canister is Benzene trap, it saves up this nasty solvent, one of the worst)
2: but do use a mityvac or HF clone hand vacuum tool , it can do wonders. I have one of these and will check all hoses



IF Evap 1, ?
i try first to see if the ECU is finding NO AIR flow, or too much. knowing what it finds, is the best clue to the solution. in every case. (but can be hard to discover this) will the scan tool tell me this? Is this where I will have the option for a freeze frame from the scan tool. these are also new territory for me. hahaha

thanks Jerry I will let you know what I find. Bill

[fixkick]

good, evap 1, far more easy, that is why i like 96 the best, the early ones are evap1,

im not sure, print out the freeze frame , for 443 failure. and we can noodle the data see if it is revealing.
the freeze frame stores the last failed DTC. if 443 then you have it stored in memory,just dump it.
it will show, all important PID parameter that are a snap shot during the failure.

other tests.
do you have vacuum at the back side of VSV purge idling hot>>? if not its blocked,
i think the valve is open at hot idle and you will have vacuum at the output of VSV too. check that.
then make sure that vacuum can suck and get air flow from the canister, this will be the tricky part., id use the vac too on the hose going down and not see it pull a vacuum.
if it did, the canister is plugged.
i wonder if it will throw codes for the hose removed to canister. vsv to canister. if not , that means the canister is plugged, up. or if you get a different DTC code. but the monitor is slow slow to trip off.

[dynabill]

good, evap 1, far more easy, that is why i like 96 the best, the early ones are evap1,

im not sure, print out the freeze frame , for 443 failure. and we can noodle the data see if it is revealing.
the freeze frame stores the last failed DTC. if 443 then you have it stored in memory,just dump it.
it will show, all important PID parameter that are a snap shot during the failure.
here is photo shots of scan tool, I am not to savvy on these yet ha! http://www.fixkick.com/bbs2/Suzuki-bbs/g...07_00.jpeg http://www.fixkick.com/bbs2/Suzuki-bbs/g...07_37.jpeg second pic is so I could get all the readings

other tests.
do you have vacuum at the back side of VSV purge idling hot>>? if not its blocked,
i think the valve is open at hot idle and you will have vacuum at the output of VSV too. check that.
then make sure that vacuum can suck and get air flow from the canister, this will be the tricky part., id use the vac too on the hose going down and not see it pull a vacuum.
if it did, the canister is plugged.
i wonder if it will throw codes for the hose removed to canister. vsv to canister. if not , that means the canister is plugged, up. or if you get a different DTC code. but the monitor is slow slow to trip off.

I will do more tests tomorrow, and let you know what I find. these are all new monitors for me, so it may take a bit to get them in my head, lol Bill

[fixkick]

those readings show , a perfect , set of readings, hot, in closed loop too.
what the tools dont do, well, is tell what parameter caused it to fail, in fact they hide it. (in this case, not others)
not:
showing, expected fuel rate or idle Duty cycle changes out of range ,would have been spiffy. but no. joy. there.
We will have to fix it by pure Guile. cunning.
we know it fails, we dont know why?
trick 1:
1: pull the hose from VSV output to Canister , small vac hose. , there is a big hose, vent to tank and small hose to VSV, pull small.
2: using new OBD tool, reset DTC.
3: start car, and see if you can make it fail, 443 or other codes. 440 and 450? with hose DISCONNECTED. some cars, you do this and the 443 are GONE forever. EVAP1 only.

4: just before that, at hot idle does it suck air? i think it must. as a hot motor, at 800 rpm parked, at 0mph, it must suck air. if not the VSV is bad, (or bad electrics to it,) is the vacuum hard vacuum nipple to it is not sucking. It must suck at all times that hose. pull it and see?
the vacuum must be there at the vsv at all times, and at idle the ECU opens this valve so both nipples are now sucking are (pulled 1 at time but not same time) ,are they. (pull input, then put back , pull output)
if it cant suck air, its bad. easy find and fix.

5: put a volt meter back probed (needles) to the VSV solenoid coil conn, do not unplug it. just probe it, with 2 leads, it must read 12vdc, (or 14.7) that means the ECU has it ACTIVATED.
is it activated a HOT IDLE>? if not, then we must find out why.? a cut wire?
if the vacuum input nipple is dead, then that line is blocked, they do. or the Vac nipple port at the source.

Id attack the VSV first.

some cars the EVAP1 tests can be fooled.
just like a 1995 EGR thermo sensors, and for the same reason.
they forget to do 2 tests, open and close, or they do the 2 tests but only fail the 2nd, if the flow is too low.
If the test is SMART it does a differential test, A -B = flow. if same, fail. my wild guess is this is what it does and we can not fool it.

but if you can fool it , that tells me, the canister is plugged up.
in other classic car forums, (they are wild) they split the casing, then pure out the activate charcoal and then put in aqurium tank carbon (walmat) and it works PERFECT (valves inside clean tool)
(this is common in calif, due to smog rules and visual inspections)
for you , i bet you just want the CEL light out. not worried about carbon block issues.

What it does and how:
what happens on a cold morning; you call in sick; the car is in the drive way, the outdoor temp goes from 70F to 110F and as the fuel expands and the fumes expand, they build nice pressure against the reg valve.
the spring tension there, is overpowered and the valve opens and BURPS, fumes into the charcoal CAN, the carbon sucks it up like a sponge.
you then get better (sick), and go for a beer, and start car, and when hot (block) the VSV opens (at stops signs) and sucks fresh air thru the CAN and burns them all, for a repeat , next hot day.
the test, only checks that air flows, from VSV nipple 1 to 2, and I'm sure nothing else.
knowing that , solves this riddle. I hope this helps in some small way.

good day to you.

you have EVAP 1
below is EVAP2
but lets talk hoses.
your VSV is on the left side, right facing engine.
SOL #21
see hose 38 , it must suck all times running, its a direct shot to the full plenum
hose 57 must suck at hot idle , does it?
on your car SOL 50 is missing and the hose is a straight shot to CAN, this is TANK VENT hose id 55

---

ps,
ive seen the plenum nipple clog ,a 5min fix. THE EGR and PCV valve make a huge nasty cloud that loves to clog vacuum parts.

[dynabill]

Hey Jerry, got up this morning and checked item 4 on your last post. I got engine to hot idle and pulled hose leading to the canister, there was no vaccuum, pulled top hose from vsv to plenum and have good vaccuum from intake, which is a good thing, lol. I them back probed the vsv and have no voltage registering from it. I reset back probes several times to insure I was getting a connection. Still no voltage registering from vsv. I now have dtc's reset and have car idleing in drive. with vsv hose detached from the evap canister. trying to get a new code to pop or the same one. per your instructions. I am going to take it for a drive with scan tool attached, it may come up with better info if a code pops.

[fixkick]

Hey Jerry, got up this morning and checked item 4 on your last post. I got engine to hot idle and pulled hose leading to the canister, there was no vacuum, pulled top hose from vsv to plenum and have good vacuum from intake, which is a good thing,


lol. I them back probed the vsv and have no voltage registering from it. I reset back probes several times to insure I was getting a connection. Still no voltage registering from vsv.
so the valve is commanded closed, i wonder why? ill look in the fSM to see the permissives,for that......


I now have dtc's reset and have car idling in drive. with vsv hose detached from the evap canister. trying to get a new code to pop or the same one. per your instructions. I am going to take it for a drive with scan tool attached, it may come up with better info if a code pops.

there are 2 conditions, one is normal, and the other is testing.
it says on page A132 the normal conditions, are hot motor, idling and driving, that valve is commanded open and sucks at the top of the canister.
the above is just normal actions, (eating fumes)
the does a test
the test rules arelive monitor rules) that trip the test off are way more complex.
coolant 158-230F
air sensor 15F to 158F (IAT sensor) im sure it is.
baro reading GREATER 75kpa (relative not absolute) 22inches HG mercury, vacuum I think they meant about 22inchs. not greater
I'm sure they are using a metric vacuum gauge here.
fuel tank above 15%
10seconds, above true

hark i found it, page 133 contradicts 132, step 3, hot engine at idle, VSV closed, then race motor and the valve opens and sucks.
use finger tip to feel this action with the hose removed.
the errors on page 132 im sure are japanese to English translations.
my guess, the at idle is closes to that you don't get funny idle , with huge fumes being sucked in, so keeps it closed. (normal ops)
but the test is done at idle in 10 seconds flat.
they , just open the valve for 10s at idle see the RPM BUMP and that bump passes the test. end test.
Toyota has some great books that explain all this 10times more clear than suzuki, and 100 times Geo books..

now you can do the hose test, at idle then 3000rpm , and you know the VSV is good. and ECU.

good luck you you.