Wankenstein

Video showing exhaust leak test. Vacuum cleaner in reverse blowing air into tail pipe + soap and water spray at location of the bottom of the manifold/donut gaskets. Used a piece of 3/4" pvc pipe to listen for leak but couldn't hear it.

https://www.youtube.com/watch?v=lImE...ature=youtu.be

ezone

Been apart and gasket reused too many times, or the flanges are slightly warped?

IDK...I really think your main problem is in the ECM tune, though I've seen a pretty darn minor leak cause a sensor fault too.

Wankenstein

New gaskets. Flanges may be warped? I rigged up a smoke tester with new, emty paint can, compressor, NPT connections and uses a cigar inside the can and it works well. I did an exhaust smoke test through tail pipe and no leaks seen. Did another test on intake side and no leaks.

I still need to back probe MAP and test fuel injectors..just sucks because of limited time (work nights) and energy. Engine doesn't stall with a/c on so, I keep that on while driving. If, after testing MAP sensor and injectors and both are good then I'm getting it tuned. I'll be purchasing a soketed ECU and the dyno shop installs whichever brand chip. I opened up the present ECU to take pics for tuner shop and didn't see any bulging capacitors..physically it looked fine (checked both sides of board).
It's needed a 28 pin socket + chip + dyno tune since the previous owner did all the swaps but, he cheaped out. If iwould have known more about the need for a tuning I wouldn't have purchased it at all or at least not the price I paid.

Wankenstein

I found this thread and it could be related: https://honda-tech.com/forums/tech-m...-idle-3092586/

ezone

I never really noticed much in the way of electrolytic capacitors with a bulge in ECMs and other electronics, but they sure did ooze electrolyte onto the board and surrounding circuits..

It used to be horrible in the 80s and 90s, you could expect many things to turn to crap at 10 years because of leaking caps. Consumer and industrial electronics included.

Most electronics seem more robust these days.


Random from a goggle image search for 'leaking capacitor'

Wankenstein

Yes..I've seen some nasty ECU board burn pics via google searches.
I might pull the ECU and inspect closer with a magnifying glass for poor solder joints.

As far as tuning it one of the options is to install the 28 pin socket, resistors, etc..(total $10) myself onto the current P75 ecu .. however, with the lean issue will that be a problem or will the tuner man just override it through chip, EPROM and dyno tune? Other option is something like this ($150): http://www.hamotorsports.com/socketedp06.html

I should have it narrowed down by this coming Monday. In that way I can give him a list of what's been ruled out when I bring the car to him Wednesday.

ezone

You need to talk to that guy.

I deal with mostly stock stuff, I never been into tuning and whatnot.

Wankenstein

Probably not related ..but, would a leak in the coolant/bypass hose that connects from cylinder head to throttle body cause poor idle?

ezone

As I see it: Only if the coolant level drops significantly, and at that point it would probably idle high and surge at times instead of idle too low.

Wankenstein

Okay. Crap weather here now but, trying to get a few more test out of the way before bringing it to tuner. Going to pull fuel injector harness one at a time with engine running to see if idle drops evenly. Also, intend to pull them all to spray test. Along with other testing.

Wankenstein

Pull vacuum off of FPR and idle raised..assume this is normal.
Pulled fuel injector harness off one at a time and the 1st cylinder did drop but not as much as the other three. I forgot to remove gas cap during drop test and while removed FPR vacuum so, when I went to remove the gas cap to relieve pressure before puling, cleaning, testing the injectors gas poured out of tank's fill hose.. also assume this is normal.

Using a 9v battery, brake cleaner, pressure to to push cleaner...I was expecting a mist while pulsing the injector however, it's just a straight stream on all four injectors. I called an injector cleaning shop and the guy told me some injectors are designed for a straight stream and not a mist. He said a 9v battery is fine and it keeps the amperage low. I am removing and cleaning the screens + forward and back flushing injectors under pulse. My DMM is not working to properly to test ohms.

When I get the injectors back in I will do another drop test and if still having an issue with just one..I will swap it to see if it causes an issue within the swapped cylinder.

Update: Injectors back on and all cylinders dropping the same now but, still existing low idle/stall issue.

When I installed the injectors I forgot to connect the fuel hose (return) that attaches to the FPR (firewall side) to metal tubing under car. It seemed like the engine ran better with that hose disconnected but. of course gas was pouring out on the ground. I also have noticed on prior occasions that when I open the gas cap no pressure escapes or at least none that I hear. It is a Honda genuine gas cap.

I made an appointment with tuner shop for Wednesday and they told me the lean idle needs to be solved before any tuning. They said it's possible a faulty transistor(s) in the ECU could cause a or stumbling lean idle.

Wankenstein

I need to try MAP testing:

ezone

Your tank is essentially open (vented to atmosphere), you have no emissions parts left to contain pressure.


The tuner dude needs to watch the fuel trims vs. O2 sensor....should be obvious the stumble happens every time the fuel trim snaps back to single digit. The lean condition is causing this to happen.
IMO.

Wankenstein

Oh yes..I should have thought that one out.


When I talked to them yesterday I expressed my concern of: if the lean issue is related to the ECU..should I purchase the re-conditioned one from HA Motorsports to have on hand if needed. He told me to hold off on purchasing it. He also said if it indeed is the existing ECU that is casuing the lean issue then it will need to be replaced with or without a tune.

The tuner mostly uses HA motor sports for their ECU needs. HA board components are tested, re-epoxied, new capacitors, metal case is powder-coated and ZIF socketed. The tuner uses/installs Crome brand chip.

Side note: I had to push pedal to the floor for a few seconds this morning to get it to start.

The following may not have any relevancy since my car's engine doesn't stall under load at idle but, I figure might as well cover as many basis as possible:

Doing some more research I found this: http://ecomodder.com/forum/showthrea...ained-967.html
Under heading 23.7.2 it states: "If the RPM dips or bobbles when a certain load is introduced, it is probably not getting an "advanced warning" about the load. For instance, if every time you turn the steering wheel the idle fluctuates, the ECM is probably not getting a signal from the PSP switch."

Referring to 23.1: Power Steering Pressure (PSP) Switch: This car did have power steering and the previous owner removed the PS pump, PS fluid and one of the PS flange tubing. I further depowered it by removing the seal inside rack.
Clutch Switch (Manual Transmissions): The clutch switch has been bypassed in this car
Gear Position Switch (Automatics): This car was an automatic..converted to manual. There is a slider switch from the automatic transmission still hooked up that the seller showed me. It's underneath the center-dash. He said if it's set in a certain postion the engine will crank but not start. I always leave it in the middle of it's range.

ezone

COULD have some bearing on certain run strategies, but shouldn't have any bearing on lean@idle issue in either switch state IMO.

Seems like you should let the tuner dude do his thing now?

Wankenstein

Appointment at 10 a.m. tomorrow morning. One hour drive. Going to have to open my wallet lol

Wankenstein

Never talk to two tuners in the same shop..will get two different replies and results lol. Initial person I talk to told me to have new (refurbished), socketed ECU on hand. Second person I spoke with said don't order ECU yet..will solder in socket if need be. Arrived this morning and the lead guy (owner) asked me if I have the new ECU and the other tuner told him that he advised me to hold off and the the socket can be added at their shop. Owner said his solder sucker no longer works and he doesn't like to solder in sockets anymore do to instability of old boards (understandable). So, bottom line no socketed board..no tune/diagnosis.

They were kind enough to adjust the idle screw so it doesn't stall anymore. They didn't disconnect the IACV to adjust idle but, it seems to be working well idles at 800 -/+ 50 no load. Around 850 rpms with a/c on. They didn't charge me at all. Seem like good guys..young and knowledgeable: build engines, fabricate exhaust and other items. Owner showed a few of their builds ranging from 250 whp to 1100 whp boosted. Coolest IMO was a boosted, street legal dune buggy.

He said a refurbished HAMS socketed ecu combined with Hondata S300 + tune is the best setup because it has more tune-able parameters, live data capabilities and blue-tooth app that is similar to the Hondash I have but with much more info. Unfortunately, in total it comes at a cost of close to $1000 and I'm not willing to spend that all at once now. So, my thought is to purchase a $10 used Denso MAP sensor part# 079800-3280 and see if that helps. If not I will consider purchasing the HAMS obd1 ecu. May revisit adjustable FPR idea as well..here's a related b20 swap thread if you care to read (reply 15 especially): https://honda-tech.com/forums/acura-...ressure-63629/
and this: http://www.jdmcity.com/archive/index.php/t-12310.html

Out of curiosity does a fuel filter have psi limitations? Say a 1.6L fuel filter in-line with a 2.0L engine.

The guys at the tuning shop complimented the seller for doing good, clean work on the swap. They also said without adding larger diameter headers and exhaust that they wouldn't expect much hp and trq gains from dyno tune.

ezone

Owner and employee have communication problems LOL
You could start with just disconnect (and plug) the vacuum line to the fuel pressure regulator for a couple days.......that will leave the pressure at regulated max all the time.....see if that changes anything. SHould boost fuel delivery a little at idle and low load, and leave delivery at high loads unchanged.


I think burst pressure will be 20x anything you can throw at it.

Wankenstein

I had a feeling that it was going to happen that way but glad it did in some regard: as in regarding my wallet..lol

Will try it.
Happen to know the stock fuel pump max psi?

Thanks. Good to know.

The Hondash scanner numbers seem to be off. RPM's on dash are around 200-300 less than scanner.
Speed is off by 5 mph.

Fuel psi at idle is 38-40 on inline gauge above fuel filter. Scanner shows 40-42 at idle.
Not sure if other parameters are off. Tomorrow before I disconnect and plug FPR vacuum I will get a reading on both with key in second position, at idle, at 3000 rpm and snapping throttle.

ezone

Pinch the fuel return line closed and watch your analog gauge. Probably near 80 PSI if the pump is good?


Should not be a scanner issue, it's displaying data exactly as the ECM sees it. I'd trust scanner data over gauges on the dash any day. Should be your 23yo gauges. You could plug in a real tach and watch RPM to confirm one or the other?

Except fuel pressure....Honda didn't give the ECM any way to monitor fuel pressure, so I have no clue how your scanner app displays that data unless it's inferred.

Wankenstein

I didn't try pinching off the return line yet. I did disconnect FPR and cap both the FPR port and TB port and idle/rpms didn't stumble or drop at all. Idling a bit high now (900-1000 rpm no load) but that's probably expected.
I fried my phone's digitizer yesterday trying to replace ($5) screen that has been cracked for a few weeks. Purchased a new phone today and need to get a new code from Hondash developer to work with the app. So, can't see what o2 and trims are doing yet.

ezone

You said ol'boy cranked the idle screw (IAB?) up a bit, so crank it in again?

Wankenstein

Will adjust IAB. I need to spend more drive time with as it seems to be behaving slightly different under load. I would assume that the ECU needs some time to get used to this change as well. It accelerates fine but seems to be just slightly rough. I should have the Hondouche app/scanner working tomorrow so I will post a video.

Why tune for $1000 when you can Ebay tune for $26 and comes in pretty, shiney colors too..lol: http://www.ebay.com/itm/92-95-CIVIC-...38.m2548.l4275

If I end up going to an adjustable I will purchase an AEM brand.

Wankenstein

Some of the rich/lean info I've read in the past few weeks talks about advancing ignition timing + higher octane fuel to combat a lean condition. I don't intend to do this but from what I understand it changes the combustion points in the cylinder. IIRC that retarding or advancing could cause cause valve damage in some engines..is this correct?

Here's some geek speak that's above my head: http://www.megamanual.com/begintuning.htm#advance

ezone

Valve damage? IMO no. Detonation damage (google image search that) would be very bad though and tear up far more than just valves.
Jacking with advancing the timing may quickly become very detrimental because (IMO) it already has a lean problem.

Wankenstein

Thanks.

I just read this: http://www.enginelabs.com/news/deton...-it-using-efi/
and this: http://www.contactmagazine.com/Issue...ineBasics.html
The b20 block on mine doesn't use a knock sensor or at least I don't see one. I never knew that fuel is used as a heat conductor within the cylinder.

Wankenstein

Gauge on top of fuel filter showing 45 psi with engine off, key in second position. Hondash scanner shows o2 sensor readings at no lower than .03-.04 volts at idle (no load) now. I will make and post a video today.

How much PSI do you know or reckon is getting to the fuel rail at idle with FPR vacuum disconnected?

I contacted hondash developer and asked him if he could include: MAP voltage, TPS voltage. Audible alaram for engine temp spikes to alert to overheating. Waiting for his reply. Some of my readings could be off due to the prior paramaters I setup previously in the Profile section: http://www.hondash.net/2016/11/car-p...-settings.html

The used, tested Denso MAP sensor should arrive soon..the current one may be fine but for $8 (free shipping) I won't be out much. Would a higher (than 87) octane fuel help in anyway?

Wankenstein

I tried to back-probe MAP sensor but can't even get a small paperclip into it.

Today using the scanner the o2 voltage did drop down to 0 volts as idle dropped/stumbled. It didn't happen much and it was less amount of time when it did...proper idle recovered quicker. If the new (used) MAP doesn't solve it I am going to purchase a refurbished ECU.

Hondash developer replied: Yes on TPS voltage and audible alarm for engine temp. Will try for voltage on MAP sensor.

I forgot to mention that '99 CR-V fuel pressure spec (per manual) is 40-47 psi. with FPR vacuum disconnected and pinched.
'92-'95 Civic fuel pressure spec: Vacuum Hose Disconnected 40-47 idle psi. Vacuum Hose Connected 30-38 idle psi.

Did some more online research and came across info that points the possibilty of a weak alternator sporadically affecting fuel pump voltage that produces dips in idle. Battery voltage on datalogging is good though. Will go to parts store to have alternator checked at various rpms. Google hits: https://www.google.com/search?site=&...k1.lmXfaZeznEM
One reference: http://www.ca.dsm.org/forums/ubbthre...&Number=171018

Reviewing datalogging from yesterday whenever o2 voltage drops to 0 volts fuel pressure reads 39-40 psi and battery range between 12.7 to 14.2 volts. o2 voltage seems to be all over the place now at idle..however, engine has never ran smoother at idle though since purchase. I do feel what seems to be a minor misfire while driving. On my drive to work today idle o2 voltage never fell to 0.

ezone

Holy cow what happened? It is supposed to be that way, now what is the fuel trim doing at this time?

Wankenstein

Just made this video an hour ago. Engine from cold start to operating temp at idle. The live data is longer than shown in video..scrolled forward to make it shorter. At 1:25 mark o2 voltage adjust more. At 2;40 mark I held down throttle to 2500 rpms or a bit.

I replaced MAP today..no noticeable change. However, haven't tried it with vacuum applied to FPR.

Will post another video one tonight