Dear All,
During test, The fuel pump was disabled, Throttle open high.
Here the results:
Cyl 1 : 186 psi ( after 4 strokes) first stroke : 145 psi
Cyl 2 : 185 psi ( after 4 strokes) first stroke : 150 psi
Cyl 3 : 190 psi ( after 4 strokes) first stroke : 155 psi
Cyl 4 : 187 psi ( after 4 strokes) first stroke : 152 psi
Ambient temp was 82F.

Based on Haynes book comp. standard: 178 psi, min 135 psi ( I don't know whether it is right or not).
It seem that the compression litle bit high probably due to carbon built up.

I'll test the vacuum for tomorrow ( too hot for today ).
I might take a video for the RPM down grade.

The engine it self runs very smooth no bang,thick,thack or any otherweird sounds.

Thank's

 

Here are the videos:
Vacuum:
http://www.youtube.com/watch?v=VpMO_METfeg
The first 2 second the idle ( around 23inHg ) it's constant ( no fluctuation ), the I rev up to 2500 put of the gas paddle suddenly. The vacuum only up around 4 inHg.

RPM:
http://www.youtube.com/watch?v=3mcax2SXO1Y
I rev up the engine then released the gas paddle suddenly.

This test was performed after engine reached its normal temp (The fan was on once ).

I seems everything normal ?

The timing belt and water pump was replaced by last owner at Honda dealer (around 8k milesago). I am just thinking the slow rev down probably due to the timing belt miss 1 tooth.
The cat converter was replaced at the same time, could be cloging exhaust ?

I justdrove the car around beach and mountains ( including labor day trafic jam at small cities, behind semi onwhen uphills )it still gets 484 miles 13 gallons (37.23 mpg), AC on 50% trip. On the other day I5 highway 280 miles for 7 gallons (40 mpg).
I seems to good to be true ??
The tiresare standard and their treads are in very good ( I thinkneed a gps).
I notice this car doesn't pull hard ( can get killedif trying toget around tripple decker during uphill ).

 

Well guys,
Finaly I traded this 99 Accord LX MT with 07 EXL V6 6SpMT.
Guess what.....on the 07I still have the same problem,rpm rev up a little bit whenI shift the gear up. So I think this ismy problem and not the car problem (since the car is new).
My feet are not suitable to Accord MT, only works for Mazda MT and Camry MT...

 

I've made a few minor edits below for improved clarity.

I think your problem may be associated with an anti pollution measure on the car.

If the airflow is not closed off fully (i.e. to idle closure) when / shortly after the butterfly valve is closed this will tend to reduce oxides of nitrogen (I think), so, some cars are set up to briefly allow some residual air flow to bypass the butterfly valve after you lift off the accelerator pedal. This will tend to mean that rpm won't drop as immediately or as quickly as would otherwise be the case when you lift off the pedal.

With Hondas this is (seems to be) achieved through the IACV (idle air control valve). You can check if this is causing your problem by disconnecting the elecrtrical connection at the IACV. You will get a CEL when you do this, but if the rpm problem disappears then the IACV is what is causing it.

The IACV is controlled by the ECU, and allows additional idle air flow primarily for maintenance of correct idle speed when things such as the air-con, lights, wipers etc are used (anything that might create a load on the engine that would otherwise cause idle speed to drop below minimum). As stated above, it also seems to be used to maintain a slightly increased airflow on the overun after the throttle has been shut off (above what would typically be required to maintain idle speed) for anti pollution reasons.

This can result in an irritating non-response (or at least lazy response) to shutting off the throttle, and can in some instances even cause a slight rpm increase as you lift of the accelelerator pedal and disengage the clutch. I had this problem on my CB7, and it used to really bug me! I thoroughly cleaned the throttle body and the IACV, to no avail.

How to fix it? I can only say what I did, which was largely effective. The simplest and most completely effective method is to disable the IACV by disconnecting the electrical connector, but you will have a constant CEL, and whenever you use the air-con, lights etc etc the engine will idle very low and vibrate. Also, on / off throttlebehaviour will become somewhat 'savage', but initial off throttle engine deceleration (i.e. braking effect) will be very noticable!At least this is the case with the f22A9 engine fitted to Australian spec CB7s, which have a higher compression ratiothat the f22A6 fitted to American CB7s. (Edit note: 'savage' is overstating it, but the on to off throttle behaviour when in lower gears will tend to be at least somewhat harsh).

You could blank off the IACV air passages with a blind gasket, which will have exactly the same affects as above but without the CEL problem.

The best compromise is to 'cripple' the IACV so it can only pass a limited amount of air, i.e. enough to allow idle compensation for auxilliary loads, but not so much as to allow the rpm to momentarily remain high (or even rise) when you lift off the accelerator. This approach will give a more refined on / off throttle behaviour and less initial engine braking effect, but if I were racing the car I'd tend to go for just disconnecting the IACV electrical connection or the blind gasket...

How to cripple the IACV? Air from the atmospheric side of the butterfly valve passes through an internal passage in the plenum casting,to and through the IACV, to the engine side of the butterfly (i.e. into the plenum chamber proper),the IACV controling the moment to moment additional air flow from atmosphere to plenum.

Between the IACV and the plenum chamber casting is a rubber gasket that seals the air passing into the IACV from theair passing out of the IACV. If you make an additional papergasket that restricts the possible airflow from atmosphere to the IACV (or from IACV to plenum) then the airflow can be tailored to permit enough airflow for idle compensation but not enough to permit the rpm problem (at least substantially reduce the problem).

There are two internal air passages in the IACV. Make the new gasket (using good quality stiffish gasket paper) so that it restricts one of these holes (i.e. the new gasket will have one large hole and one small hole, the small hole governing the max quantity of air that can pass through the IACV). I can't remember the size of the small hole (in gasket) that I ended up using on my car, but about 5mm diameter will be ballpark, but you will need to experiment, the aim being the smallest hole that will still allow adequate idle speed compensation when air-con etc are in use.

A very small change in the diameter of this hole will have a significant affect on the amount of air that can pass through it. It's clumsy to drill and file neat holes to size in cardboard, Imade thehole (holes, including my experimental ones) using a leather hole punch, the type that has a 'starwheel' of altenative hole punch sizes. If you want / need a hole that is in between sizes on the leather punch, you can 'nibble' the edge of the hole with the punch.

On my car I've set it up so the idle speed will drop slightly with air-con on, and vibratea bit more than is usual, though not if I'm only using the lights and the wipers (rarely need the air-con at night, which is just as well since the loads are cumulative). This is my choice, I could easily fix this, but if I increase gasket hole size slightly to address it then therpm problem will slightly increase.

Be aware that this 'fix' will affect pollution control effectiveness slightly, so if you ever need to pass any sort of 'smog' test it might cause 'issues'. Experimenting with the gasket hole size is easy since (at least on my CB7) the IACV is easily accesible / removable, and you can swap gaskets over in just a few minutes. Once you find the gasket hole size you're happy with, make an aluminium restrictor plate to the same dimensions as the gasket and ditch the gasket, because over time the unsupported parts of the paper gasket (around the smaller hole) may deteriorate and be sucked into the IACV and / or the plenum.