Civic '92 throttle response delay, bad idle

Few days ago i replaced complete DPFI throttle body (the one with two
injectors) with "new" junkyard part. The old one had worn throttle
plate shaft that was sticking like hell and making my life miserable
in stop and go traffic.

Well the sticking is now gone but there are new problems that make the
previous one a joke :-(

1. After car is warmed up there's throttle response delay so if I
don't apply throttle really really gently the engine will first stalls
for about 0,7sec before revving up.

2. After engine breaking and pressing the clutch engine almost stalls,
most of the time it recovers by itself and idles fine however idle is
a little bit too high or low.

3. When cold, engine starts a little too fast. With old throttle body
it always took about 1 sec for engine to fire up. Now it starts almost
immediately after turning the key. Should it be like that?


As far as I checked there seems to be some vacuum issues. The upper
tandem throttle plate is not opening under throttle and I can't feel
any vacuum on it's diaphragm hose, not even on air intake chamber
which it is connected to.

After disconnecting EACV engine RPM drops to around 400 rpm and I can
still feel a little bit of vacuum. Not much but it's there.

As I understand there's a coolant line that goes into throttle body
(not sure), should the system be bleed after TB replacement?

I also checked throttle angle sensor and it seems fine.

Thanks for any help!

 

check tandem valve and tps.

braking, not breaking. unless you mean your engine broke. [sorry, but
you're getting the flak someone else started - please get the right one.]

set the base idle correctly - if it's too far out, the eacv takes too
long to kick in.

instant start is unusual - could be an injector leak. run injector
cleaner through it.

it's switched by a solenoid - so no vacuum all the time. check hoses
and diaphragm for leaks. run motor with air cleaner off - you can see
if the valve is operating at higher revs.

also, some dpfi's have a restricted tandem valve - check to see if it
opens all the way and whether that matches your old one.

self bleeding.

how'd you check?

any codes? and are you in the usa? dpfi was only on the 88-91 civics
over here.

 

check tandem valve and tps.

Tandem is not working that's for sure.

Sorry for my English, haven't used it for a while, you can expect more
nonsense to follow

Is this the procedure?
http://media.honda.co.uk/car/owner/media/manuals/ConcertoManual/62sk301/6-133.pdf

I have two good injectors on old TB that I can salvage. Can injectors
be removed without damaging the seals since there's no way I can get
new ones?

Diaphragm is ok, I have checked it with small vacuum pump, it doesn't
leak. Solenoid is open when engine is off so I guess it's also ok.
Bloody thing just won't move.

It seems they're only two possibilities. There's leak through seal
between air intake chamber and throttle body (almost impossible) or a
leak through EACV that is also unlikely since it worked just fine with
old TB

I'll try connecting old TB to a vacuum hose to see if it works any
better.

That crossed my mind so I already checked that. They're the same, I
can open it manually all the way.

I kinda hoped that would be it.

Not by the book to be honest, just measured resistance. I didn't find
any spikes, readings are linear as I move the throttle.

No, no codes at all.

I'm from Croatia and the Civic is for French market. Most of the 5th
gen. Civic's here are DPFI, 3rd world country specs

Thanks a lot for your help!

 

did you check the electrical for the solenoid? does it click if you
apply direct voltage? is the connection good?

yes. check the idle rpm's for your car - they vary a little from model
to model.

it's not so easy to do without damage. assume for now that the
injectors are good - honda injectors are /very/ reliable. simply put
injector cleaner in the tank when you get this thing running properly.

check for damage or obstruction. if you remove the air ducting, you
should be able to operate the valve with your finger. once moving, look
for gotchas like linkages. also, the rubber hose between the solenoid
and the tandem valve diaphragm is frequently shot, so i recommend
replacing it.

set the idle correctly for the new tb. makes a huge difference.

ok, good. at high mileage, the tps can wear, but it usually sets a
code. they can be repaired, but it's a pita. easier to replace.

 

Tandem is not working that's for sure.

I tried following for today:

Connecting old throttle body tandem diaphragm directly to air intake
chamber (using new hose) of running engine. Almost no movement
regardless RPM.

Connected current tandem diaphragm directly to air intake chamber
bypassing solenoid. Same results as above.

Checked air intake chamber rubber seal - undamaged.

Checked voltage on solenoid - it's there

Disconnected EACV from intake chamber (thought it's stuck open) and
plugging the hole on it - still no movement of tandem diaphragm.
Interesting thing is that I still feel some vacuum on EACV even with
engine above 3krpm. Shouldn't EACV be fully closed when engine is not
idling?

I also set base idle. The manual states that when you turn lights,
defroster etc. idle should stay the same. Mine does not but drops down
with everything turned on for almost 200 rpm.

So any ideas what's happening with tandem? Is it possible that
diaphragm membranes on both became too hard? They're holding vacuum
and I can quite easily move diaphragm shaft. Or there's simply no
vacuum in air intake chamber. I just can't see where it could escape?

 

that's the high pressure side, so no dice. the low pressure side, where
the solenoid venturi is, is internal. the second pipe you see is just
venting for when the diaphragm is released.

ok, let's ignore the eacv for the time being.

if the tandem valve is not moving, it can only be the following:

1. the linkage/mechanism is jammed. but you say it's free to move.
2. the diaphragm is leaking. re-test by removing the hose, wetting a
finger, then with the throttle plate in the open position, put your
finger over the diaphragm pipe hole. if the plate stays open, the
diaphragm's good. remove finger, the plate should snap shut.
3. the diaphragm hose is leaking.
4. the solenoid's not working. you tested the electrical, but do you
hear it clicking if you apply current to it? if you attach hose and
blow, can you feel it open when you apply current?
5. there's a blockage on the solenoid venturi in the throttle body.

 

Connected current tandem diaphragm directly to air intake chamber

How could I be so stupid... sorry for wasting your time.

When I took off solenoid I thought that it's connected to fuel line
and that this port is only used for lubrication since it had little
fuel in it. I guess it came in there during disassembly of TB. A
friend of mine which is professional mechanic for heavy equipment
confirmed that... but he was wrong.

Thanks a lot! I'm now almost sure solenoid is stuck closed. I'll
replace it with one from old TB first thing in a morning and let you
know if this fixed the problem.

 

1. the linkage/mechanism is jammed. but you say it's free to move.

I tried replacing solenoid today. Of course one of the screws is
seized and I had to take whole throttle body down but still didn't
managed to do it. This will have to wait for tomorrow.

On port on the lower side of the TB was totally blocked by liquid
gasket. I know it's not tandem vacuum port but is it used for anything
else?

This was the plugged port:
http://kerozin.dyndns.org/tb.jpg

Also is it possible to fry solenoid by reversing polarity? Tried to
check old one first on one side and since nothing happened reversed
polarity - still nothing. I had voltage passing through it but it
didn't click and current was something hardly measurable (50mA?).
Someone mentioned some kind of diode in it. I totally dismantled it
but can't find any. Now when I apply current directly to electromagnet
it works just fine.

 

bingo! never use liquid gasket dude - this a textbook example of why!
either make a new gasket out of an old envelope [if you're desperate] or
buy the real deal.

probably diode protected, but don't worry about that - just worry about
whether it clicks and opens when the correct polarity is applied. and
don't worry any more about taking it off - you've found your problem.
unplug the ports and everything will now work.

 

bingo! never use liquid gasket dude - this a textbook example of why!

Actually that was a work of professional. It's quite hard to obtain
parts for older Hondas here in Croatia so improvisation is quite
common even for dealership workshops. By the manual there should go O-
ring which is of course not there so I'll have to use liquid gasket
again.

I've just checked and plugged port is actually used by MAP sensor.
Unless ECU won't open tandem without high vacuum reading from MAP
sensor I would still presume dead solenoid. i just hope it's working
since I butchered old one really really bad. It's made from one piece
of injected plastic, had to use dremel tool to take it apart.

I'll put tomorrow everything back together and let you know if problem
is fixed. Thanks again, without your help I couldn't do a thing.

 

understand. you can buy parts online from the usa though. if they
won't sell you, i expect one of us can buy and ship to you instead.

not ideal, but if you have to, this time let it fully cure before
reassembly. otherwise it'll just squeeze out and block things again.

it's a while since i've taken one of these apart, but the tandem
solenoid /definitely/ gets its vacuum from low in the throttle body, not
high where the vent pipe goes. i'll check and reconfirm from the manual
when i get home.

didn't know that, but it makes sense.

no, presume it's not getting signal if the map sensor venturi is blocked.

between the two bodies you have, you should be able to assemble one
complete unit that works.

 

it's a while since i've taken one of these apart, but the tandem

Bad luck. Both solenoids from "new" and old TB are blown. I'm right
now trying to repair one of them but it'll be really hard to do so.

Btw, tandem vacuum port is actually between two plates, right in the
level with solenoid.

This is what I have to put back together:
http://kerozin.dyndns.org/DSC00877.jpg

Another bad thing is that I'm reading unusually high voltage on ECU
side of solenoid plug. It's over 17 volts and once engine is fully
warmed it's constantly there regardless RPM. Blown ECU would really
really depress me.

I'm gluing with araldite epoxy one of solenoids, so tomorrow after
soldering back coil and diode we'll have another try.

 

slight problem!

it's hard to see how you can get 17v on a nominally 12v system. and
it's /highly/ unusual [but frequently misdiagnosed] for an ecu to die.

you may save yourself the time and effort by just getting another
throttle body. even if you get that thing back together, it'll be
highly suspect. shouldn't really disassemble a solenoid, just replace.n
only reason to try repair is if parts are otherwise absolutely
unavailable.

 

Well to be honest I quite enjoyed it despite hours of time wasted.

And guess what - IT WORKS! This is finished product:
http://kerozin.dyndns.org/DSC00878.jpg

I glued the thing back together, threw in 1N4005 diode for coil surge
protection (since I couldn't find old one in solenoid) and tandem is
now working. I'll have to get new solenoid since this one will fail
for sure after some time since it's no longer watertight and I have
doubts how well it holds vacuum.

The question is why two solenoids failed, wouldn't be surprised that
something is blowing them. We'll see over period of few days.

And the car is so much faster now - manual tranny, everything stock
minus rear seats, spare tire and other nonsense... not bad for
something rescued from junkyard.

One the other hand I still have what looks like EACV issues that I
mentioned before. Idle w/o load - 850pm. After turning on lights, rear
defroster etc. - 500 rpm. Shouldn't EACV take car of such things? When
I disconnect EACV connector RPM drops to around 700 rpm w/o load which
should be OK?

 

excellent!!! impressive too - that solenoid was munged.

i doubt they did - just that the venturi was blocked.

yes it should. what are the rpm's when starting from cold?

does it click when you disconnect it? you should hear it clicking as
you take the plug in and out. and revs should change. if no clicking,
the eacv or the electrical connection is shot. if it clicks, take it
off and see if the filter screen is clogged. i removed the screen
completely on my 90 since 89 and earlier civic's don't have one at all -
i figure the 90 can run without.

 

yes it should. what are the rpm's when starting from cold?

Just a tad over 1500 rpm.

I just noticed that idle is not always the same. On every stop I get
different idle rpm's which are stable but can be anywhere between 550
and 850 rpm.

Will try that in a few days and I'll send injectors from old TB for
rebuilt since I can get that done for free.

 

sounds like insufficient air flow on the eacv valve - you're idling
without supplemental control.

bet you never need to use them!

 

sounds like insufficient air flow on the eacv valve - you're idling

Damn, drove a car to work today and throttle response delay is still
present and I can fell slight hesitation around 3krpm but that was
there before TB swap.

I can't think of anything on TB that may cause such behavior. There's
fuel pressure regulator on it and nothing else.

I did change spark plugs and fuel filter at the same time as TB but
engine is working smoothly so I don't think that has anything to do
with it.

 

check/replace plug leads and distributor cap/rotor. also, be sure the
timing belt is tensioned correctly. if it's slack, there's a lot of
flutter in the timing signals to the ecu and it starts "getting
conservative" with ignition timing. [do /not/ overtension the timing
belt either! google this group for instructions or follow the honda
service manual instructions to the letter.]

 

check/replace plug leads and distributor cap/rotor. also, be sure the

I still think that reason for throttle response delay and those
strange instant starts from time to time lies in the "new" TB since
those symptoms where simply not there with old one.

So I replaced today fuel pressure regulator - nothing has changed.
Tried revving w/o lambda sensor, delay still present.

Since this is really starting to piss me off, I'll recheck throttle
angle sensor, replace injectors with old ones and well, after that
there's nothing left to repair on TB.

Will post results tomorrow.