VVT / Closed loop problem on my Silvertop.

[The Bloody Panda]

I'm running a Silvertop on a Blacktop ECU.

When I unplug my VVT solenoid on my Silvertop my engine runs perfect. With the solenoid plugged in I get severe hesitation, stumbling, and the AFR is very lean when I open up the throttle half way or go through the range of RPM in which VVT operates, but above or below that RPM range the engine runs fine. The car is undriveable with the VVT plugged in, the acceleration is too slow to turn on to high speed roads when cars are way off in the distance.

So far I've replaced the VVT solenoid and swapped ECUs, both ECUs looked good inside. No difference in symptoms what so ever with either of these changes.

I'm seriously considering changing the VVT intake cam pulley now since I'm out of ideas but I don't want to drop the money on it since everyone everywhere is saying that the rattly noise that these things start making doesn't affect performance. Well mine rattles, it has rattled ever since I bought the car 10 years ago so I don't know how long it has been rattling.

Does anyone understand the way the intake pulley works enough to tell me whether or not it can affect VVT performance in a way similar to what I'm experiencing?

I'm ready to throw down the cash for the pulley, I just need an expert opinion to tell me that it could possibly fix my issue.

[jondee86]

I'm running a Silvertop on a Blacktop ECU.

Just taking a punt here, but this ^^^ could be a big part of the problem.
The ST VVT is a simple on/off setup where the ECU triggers the VVT at a
specific rpm. But the BT VVT can trigger over a range of engine speeds
according to load. Take a look here... http://www.billzilla.org/vvtvtec.htm
for a better decription of the ST vs BT VVT operation.

Unplugging the VVT means the VVT stays in the OFF position, and this will
give satisfactory operation across the whole engine speed range. The VVT
"kick" will be missing, but otherwise the car will drive fine. However, if the
BT ECU triggers VVT early, the inlet cam advances and the inlet valves open
and close earlier. At low rpm this will mess with your cylinder filling, and
cause loss of power. At high rpm the engine should run fine.

So the questions are... how long have you been running the BT ECU and
has the problem always been there ?

Cheers... jondee86

[The Bloody Panda]

Good questions. I have been running the Blacktop ECU for 6 years but only dealing with this problem for around 1.5 years. I've owned the car for 10 years but the first 4 years it never left the garage. When I got it running finally I started out on Blacktop electronics so I could get rid of the AFM and run velocity stacks.

I was never quite sure if the VVT was working at all. It seems to me that over the years the VVT went from barely noticeable, to not really there, to making my car run lean/worse, to the terribleness that it is now. I started noticing the engine actually running worse with VVT than without 1.5 years ago.

That article on Bill's page explains why my car would run best when it was cold, since the Blacktop ECU won't activate the VVT until a certain temp is reached. Thank you for that link, now I know why that was happening rather than it seeming erratic to me.

I still just don't understand the VVT cam gear enough to determine whether or not it could throw my timing off to the point that it would make my car run terrible when VVT activates? Or if the engine would just simply cease to have the benefit of VVT due to a worn intake cam pulley?

[yoshimitsuspeed]

There is a hesitation thread in the Toyota Corolla AE86 section that you should look up.
There have been a number of people with similar problems and a number have found the car runs better with the VVT disconnected.
We can't be sure everyone is having the same problem but many of their symptoms sound similar to yours.
It is good to hear you haven't always had this problem on your setup. Running the ST on the BT ECU shouldn't cause you a problem here because the ECU is expecting the change it's implementing. It's not that the ST should have it's original timing it's just that the BT had a more advanced method of controlling it. This definitely (theoretically) shouldn't be an issue.

Going more off the other comments in the other threads and listening to others with similar symptoms I have a hard time believing it's actually the VVT or directly related components. I don't really have any helpful input on what it could be though.
Whatever the case it may be helpful if you can find that thread and read the suggestions there and see if your experience correlates closely with theirs.

[jondee86]

If the engine runs fine with the VVT solenoid unplugged, but runs bad with
it plugged in, then the VVT is causing the problem. And since this is quite
an interesting subject, I decided to do a bit of digging, and turned up this
quite good explanation of how VVt works, and the differences between
the ST and BT versions...

VVT: VVT is found on the 1st and 2nd generation 4AGE20v. VVT advances
the cam 15 degrees on activation. This means once ON the cam ratchets
forward 15degrees. The Silvertop engine (gen1) VVT comes ON at 4400rpm
and is OFF at 6000rpm. On the Blacktop (gen2) VVT is dynamic. It operates
on several parameters of which the two major ones are throttle and rpm.
Once past 60% throttle VVT kicks in. The operational rpm range for VVT is
3000rpm to 6000rpm, however VVT has a dynamic range of 1600rpm to
6500rpm. It will differ due to too many variables to list, but these are the
peaks at which it operates. The car may opt to end the VVT cycle at 5500rpm
or extend it to 6200rpm, just depends. It has a matrix in the stock ECU it
uses to compute. The reason VVT ratchets back at 6000rpm+ is because
the valve overlap moves from short (torque creation) to wide (max flow).

Several years ago, I decided to map the 4AGE20v in my country.The BZR
was tested for the first two states. ON, as stock control, and OFF, solenoid
disconnected. The graph showed that with VVT ON peak hp was less than
when OFF, but the torque suffered greatly when OFF. This was to be
expected because VVT is for torque. Even more clear was the difference
in the overall powerband. It wasn't until after 6000rpm the OFF figures
started to climb on the ON figures.

And since I often get the effects of advancing and retarding cams mixed
up, I had a look for something on that as well, and found this very useful
diagram (thanks Jeff)....



This makes it clear that with the VVT in the OFF position, the intake cam
is retarded for reduced overlap and better torque. And in the ON position
the cam is advanced for better power at high rpm. This pretty much
validates the comments I made earlier, but does not throw a lot of light on
the problem. Might be worth taking a look at your cam timing just in case
the belts has jumped a tooth.

Cheers... jondee86

[yoshimitsuspeed]

This makes it clear that with the VVT in the OFF position, the intake cam
is retarded for reduced overlap and better torque. And in the ON position
the cam is advanced for better power at high rpm. This pretty much
validates the comments I made earlier, but does not throw a lot of light on
the problem. Might be worth taking a look at your cam timing just in case
the belts has jumped a tooth.

Cheers... jondee86

It's not quite that simple.
You will notice in the description you posted that the VVT is turned off in the higher RPM.
This can be confusing because overlap is more important in the higher RPM. Unfortunately more important than that is the cam staying open long enough. If you could control lift and duration you would have it open earlier and close later but since that is not an option it's better to have it close later.

So in low RPM it's off to keep the idle steady and maximize low end.
In the mid RPM it's open where the overlap helps increase the airflow through the motor.
Then in high RPM it goes back to low overlap so the cam stays open later.

Now I say I don't believe it's the VVT it's self related to other peoples tests because some of them have run the car with the VVT triggered on the whole time and they didn't see the issue then.
This is a test I would recommend T-B-P just to see if your experience parallels.
I don't really know what this tells us but it does make me think that it's not a mechanical issue with the VVT.
Perhaps the VVT is causing a draw in the ECU creating the problem or maybe something else strange related to the VVT.
I don't know but I have been watching this other thread and hearing about people with similar issues and I have spent some time thinking about it but I can't think of what it could be. Especially since it sounds like it's getting more common.
Either that or it may be that the bogging issue that has plagued the 20v ECUs for decades is getting better diagnosed and better described but it may still be that same age old issue.

[BryceWSJ]

i believe it's due to the open trumpets and it's as simple as that. i bet if you made a standard airbox and ran a filter on the end, it would almost correct the problem. the theory is too much air is being sucked in too quickly. thus making it lean. especially when VVT kicks in, obviously you have that much more air being drawn in. the only real problem might be is the open stacks and the ECU doesn't know what's going on.

[The Bloody Panda]

Well Club4ag obviously isn't notifying me via email when threads I've subscribed to get posted to or when I receive private messages. I thought everyone one had forgotten about me, but it looks like I'm still special.

I'm going to test out the fuel system tomorrow. Thus far I've been overly focused on the VVT system, all the ignition components, and checking for vacuum leaks. Considering what I'm experiencing, and the new ways in which I'm interpreting my running condition due to all the research I've been doing, I really feel like it's going to be the fuel system. Besides, I've already been through spark, air, and swapped ECUs as well as some sensors.

I've read so many dead end threads from years gone by that it really pisses me off. I'll be sure to update this thread if/when I find a fix.

[The Bloody Panda]

***First of all, I forgot to mention in my previous post that this issue has gotten even worse and now I'm leaving the VVT unplugged for all testing and still having issues. If I plug the VVT in, the car runs twice as bad at half throttle as it does when VVT isn't connected.***


Tested fuel system today and there was no luck. The fuel pressure is slightly high, it's suppose to be 28 at idle but mine is 34. With FP and +B jumped it should be 34-40 and mine is 42. WIth FPR disconnected from vacuum it should be 34-40 and mine is 42. When I turn off the engine pressure drops to 38 and seems to hang out there indefinitely. So I feel like my FPR, fuel pump, fuel filter, and fuel lines are good. Maybe the FPR is in slight need of replacement since I am getting a pressure that is slightly out of spec, or maybe it's that Supra fuel pump in my gas tank. Either way, the fuel pressure is always slightly too high so I should be running rich.

If I cruise the AFR is fine, if I open up the throttle half way it instantly goes off the scale lean, if I then go ahead and floor it the AFR will instantly richen up into the 13's. This is quite confusing...

Well even though pressure is good I still can't rule out fuel injectors. I don't know enough about the fuel injectors to say that they're good even if the AFR is fine at WOT. My car use to run perfectly fine on my current setup for several years, but now even when the AFR isn't lean it's still not as rich across the board as it use to always be. Sometimes when I go WOT the AFR will hang out at like 15, so that's not ideal, but the times when that happens are erratic.

I feel like I could tune around this with a stand alone but that the problem would be hidden underneath that tuning.

I would think I had a vacuum leak if my car wasn't idling at 880 rpm, which is exactly what it should idle at according to the FSM.

So I guess I'll test the injectors and send them off to Witch Hunter Performance to get cleaned if nobody else has any other theories for me to test.

[yoshimitsuspeed]

This really sounds like a closed loop problem which should really mean O2 sensor, wiring, grounds or ECU.
I forget if you said but have you done anything with your O2, tested it or tried a new one?
What O2 are you running?
A lot of people have had issues running anything other than OEM denso.

[The Bloody Panda]

Ohh, well that's interesting about the problem with anything other than OEM Denso. I've swapped O2 sensors twice now since this problem has occurred. Neither is the OEM O2 sensor though, but both are 4 wire Denso models. I hate to have to dump yet even more money into the O2 sensor department but I'll do it if anything other than OEM won't work.

It looks to me like the OEM Blacktop O2 sensor is Toyota part number 89465-19685. It doesn't look like Lithia can get it though, or rather it is at least not something you can simply order off their website. Do you know what the Denso part number is? Or where I can find 89465-19685?

I had read some stuff that made me think the readings that 4 wire O2 sensors give are pretty generic and that any sensor would pretty much work. I felt further convinced of this by the universal O2 sensor options that are available as well from Bosch and Denso for DIYers. So I figured the only differences lied in the wire length, plastic connector, and mounting style.

I agree though that this does sound totally like a closed loop problem. The very first thing I did to correct the problem was actually swap O2 sensors since my AFR just sucks at partial throttle and at WOT isn't so bad since the O2 is ignored. However, even WOT is going slightly lean now. Does the ECU modify the open loop fuel map after it learns about how to run the car in closed loop?

I just sent my injectors off to get cleaned since when I have an O2 sensor monitoring only cylinder 4 the engine runs noticeably more lean than when I have an O2 sensor monitoring all 4 cylinders. However, the two O2 sensors that I used in the separate locations are different models from Denso. One is for a Toyota 4Runner and the other for a Subaru Impreza. So this either implies that the fuel injectors are not spraying equally or that the O2 sensors are both giving the ECU incorrect information but that one is more incorrect than the other. Although the O2 sensor that was monitoring all 4 cylinders was throwing a code 21 so maybe I can't use logic so much. I had modified it to have the heater always on since I had heard it wouldn't run as lean that way. I never actually tried out the O2 sensor without the wiring that way since I had to modify the wires to make it reach and have the plastic plug work in the first place.

The two O2 sensors that I've been running are Denso 234-4162 and Denso 234-9011.

Do you have a link where I can read about others having issues with their non-OEM O2 sensors? I just need a little more motivation to buy oxygen sensor number 3.

UPDATE:
Just checked the resistance on Denso O2 sensor 234-9011, which is the same one that Twos R Us is selling as a first time fit sensor for the 20V, and the resistance is like 0.9 at room temp. Quite a bit off base from the 11-16 that the FSM is calling for.

This still doesn't explain to me though why my car still runs lean at WOT, not as lean, but still lean. The more I drive the car though, the more lean it gets at WOT. Does the ECU reconfigure open loop mode after learning about how to run the engine in closed loop mode? In which case, this would explain what I'm seeing. It would even explain why my car runs extra horrible when VVT engages. Come to think of it, when I first started seeing this problem a year and a half ago it was shortly after I swapped in my old and first Denso 234-9011.

I'm going to rewire the Toyota 4Runner O2 sensor that I have and unmodify it so that the ECU can control the heater on it rather than it always being on. That should definitely get rid of the code 21 that it throws. The resistance on Denso 234-4162 is 13 at room temp. This sensor is incredibly cheap though compared to the OEM sensor, is there another difference beyond the resistance that I need to be aware of when sourcing a substitute for the OEM sensor?

[yoshimitsuspeed]

Logically I have to completely agree with you about O2 senors. They are very basic units and i believe Bosh and other quality manufacturers should have the theory down pretty dang good.
I am just conveying info that I have read from other older posts on the traditional 20v bogging syndromes.
I also couldn't pull up a thread any quicker than you. I believe one was on MR2OC. I believe another was on here. Both were a couple years ago. My saying OEM is kind of shorthand. Due to it being so long ago I don't remember all the exact details. If I remember right one person tried a couple Bosh units with no success but swapped in a junk yard Toyota O2 sensor from another motor and claimed it fixed the problem. Others claimed that Denso OEM equivalent seemed to fix their problem.
While it seems extremely unlikely and I still believe a proper Bosch or similar quality sensor should be a perfectly good replacement I have a similar story that I can't completely explain.

After I swapped my 20v into my AW11 and drove it without problems for months without issues I started getting strange AFR issues. I was running a fairly new Bosch O2. I think I got a cheaper 4 wire for another vehicle since I figured they worked the same. I should note that I have heard that 4 wire sensors are designed to be a certain distance from the head so while they should all operate on the same principle they may heat differently based off the temp they should see for their designed location. This is the only variance I know or at least have heard is a possibility among common 4 wire narrowband sensors. I should also note for anyone who hasn't followed my builds this is running my BT turbo on a GZE ECU. Anyway I ran this senor for some time without trouble. Then all of a sudden I notice my cruising AFRs start to wander off from time to time. This isn't as predictable or regular as these 20v reports have been. Everything would be working fine then I'd be driving down the highway and all of a sudden my AFRs would go from stoich to reading very lean or on occasion very rich. I went through all the troubleshooting and diagnostics and couldn't find any problems. The O2 passed all tests per the FSM.
I could tell by the behavior though that this had to be O2 related. I ordered another Bosch again assuming they were a high quality sensor (I still believe this). I think I ordered a GZE sensor at this point in case it was because I got the wrong sensor originally. I swapped the sensors and the really weird thing is the behavior changed but still didn't act normal. This was a long time ago so my story may be a little off but I do remember the behavior changed. As I remember it the intermittent behavior changed. May have started going way rich more often than lean or something like that. It also started getting weird when I pulled up to a stop. When before AFRs did exactly what they were supposed to do now they would go so rich or so lean the motor would almost stall. On a couple occasions it did. Thank god I had my wideband and could study the behavior. I went through everything. I made sure the O2 was hooked up right, made sure the wiring was right and healthy, made sure the shielding was grounded to the right location, posted threads and talked to all the experts who could help me. Nothing had any effect. Nothing showed any errors or flaws. I could not figure out what was causing the problem.
Since changing the O2 sensor changed the behavior I was pretty confident it was still related to the O2 circuit. I was running an AFX wideband which is an awesome unit but it doesn't emulate the O2. I decided that I would much rather buy a wideband that could emulate than try another narrowband. I bought an LC1 and wired it in to emulate the NB. It took some tuning to get it to recognize it and get it all reading right but once I did everything worked great. The problem disappeared. The ECU looks for a particular signal to switch to closed loop on startup and I didn't spend enough time trying to see if I could get the LC1 to emulate this signal so now on startup I need to blip the throttle once to get an AFR sweep and then it kicks into closed loop. After that it has always worked great.

If your wideband has the abilty to emulate maybe try plugging it into your O2 input on the ECU and see what happens. IMO it would be a much better route than dropping big money on trying to find the exact OEM sensor.

[The Bloody Panda]

I rewired the Toyota 4Runner sensor so that the ECU has control over the heater circuit so I'm going to go ahead and try that out before anything else since it's free and the most straight forward to do another interesting test with. If ever I was on the brink of fixing this issue now would be the time since it's just my luck that I sent my fuel injectors across the country to get cleaned about 4 hours before I read your post.

I've been thinking more and more about it and I am quite sure that all these issues started after I ruined my Bosch O2 sensor and replaced it with that sensor that Two R Us is selling. The process of it running leaner and leaner was gradual, maybe the ECU doesn't make drastic changes to closed loop even if the new O2 sensor was telling it to, but over time it eventually started taking the information that the O2 was giving it and following it exactly. That's the theory anyway.

It was late 2012 when I swapped that questionable O2 in and ultimately ended up unplugging my TPS so that the ECU would force the engine to run rich. I was out of money and time and ended up putting the car on the shelf that winter so I never really got around to diagnosing the problem until now. Your questions and incites have really helped to jog my memory though, thanks for the help.

I imagine I'll get my fuel injectors back in a week and a half so there won't be anymore progress until then. I look forward to seeing the test results on the injectors but I doubt they were in bad shape. The injectors looked brand new other than the fuel on them.

[yoshimitsuspeed]

I did forget to answer your question about open loop. Both the NA and GZE 16 valves are able to change open loop trims based off of closed loop inputs so I would be quite sure the 20 valves will too. So yeah, if the O2 is giving bad readings it could cause your open loop AFRs to change over time. If this is the case an ECU reset should eliminate or change the behavior.

[The Bloody Panda]

Alright, I unplugged all the connectors from the ECU today and it'll be unplugged for over a week while I wait for my fuel injectors to return. I'll first see if I can run the engine without going lean with this Denso O2 sensor that has proper resistances according to the FSM. If that actually works, I'll plug the VVT back in and see what happens to my AFR. I didn't want to get my hopes up, but I really feel like this could be the fix.

[The Bloody Panda]

Got my fuel injectors back but the problem still continues with the Toyota 4Runner O2 sensor in place. I have no clue why this thing is running lean in closed loop. I feel like a stand alone ECU would solve this problem. The fuel is available since the pressure is good and so are the injectors. The ignition system has been serviced and I feel like it's working 100%. I can't find a vacuum leak anywhere and my idle is smooth and steady at 880RPM.

I tried reducing the air coming in by adding restrictive filters to my velocity stacks and blocking off my cold air intake but this didn't improve the lean condition at all. Besides, the factory Blacktop ECU should make a Silvertop run rich, adding velocity stacks probably increases the air flow enough to make the Silvertop flow air like a Blacktop with the air box. Either way, the ECU should be able to compensate for the increased air intake by monitoring it with the MAP sensor and air temp sensor. After all, this combination I've got going of a Silvertop engine and Blacktop ECU use to make my engine run slightly rich all the time.

Anybody got any suggestions for me to try out? I feel like I've already done most everything to find a solution to this problem. Currently I'm looking to find a plug and play ECU for the 20V. I know Battle Garage is working on one but I don't know when it will be available.

[yoshimitsuspeed]

What happens when you disconnect the TPS?
Have you tried testing it including a sweep test? See if you can find any spots where the readings jump unexpectedly. No idea why it would change with VVT but I don't have a whole lot of other ideas.
Do you have NB emulation on your wideband? If so all you would have to do is swap out the O2 signal wire for the WB signal wire. Leave the heater on the NB hooked up so it doesn't throw any codes.

Have you monitored the NB output and input going into the ECU?
This would tell you a lot especially if you can monitor it while you are seeing the leanout condition. What I have done is plugged a wire into the O2 signal port in the diagnostic box and run the wire through the door into the cabin. From there I hooked up my voltmeter in the cabin and was able to watch the voltage. You could even get one of those cheap NB gauges and wire it in. Now you should be able to see what's going on. If the voltage reads normal when the wideband is reading lean then something is wrong with the sensor or the wiring up to that point. If the NB is reading lean and the O2 isn't compensating then either the ECU isn't reading it properly or there is a problem between there and the ECU in the wiring.
You could also tap into the O2 signal at the back of the ECU to make sure the signal looks good all the way up to there. If the signal looks good up to the ECU then I would suspect the ECU. If there isn't something wrong with the ECU then about the only other option would be something throwing it into open loop but that should cause a CEL.
Does your CEL blink normal when you jump the pins?

[burdickjp]

Open stacks won't do this. Don't worry about that. Unless you're running really long or really short stacks they won't be a problem, and even if they were, it wouldn't be this much of a problem.

Have you checked ECU codes?

To me this sounds like a sensor problem. Try disconnecting the MAP, TPS, or both.

The advantage of a programmable EMS is that you can look at exactly what your sensors are telling you and very quickly deduce when they're screwed. The stock ECU only says so much. I've honestly only seen a single 20v run awesomely on the stock ECU, and that was my first blacktop. It was a good thing, too, because it gave me something to aspire to with the rest of them. I had, unfortunately, sold that stock ECU. I later went megasquirt and haven't looked back.

[jondee86]

How (exactly) is your MAP sensor getting a signal ? To what and where did
you hook it up ? I've read thru this thread, but I haven't picked that up.

Cheers... jondee86

[The Bloody Panda]

First of all, thank you very much for all of this great help. I've got some work cut out for me thanks to all these excellent questions.

What happens when you disconnect the TPS?

It runs excellent at WOT but throttle response is horrible of course for partial throttle. It also runs off the scale rich when cruising.

Have you tried testing it including a sweep test?

No I have not, I don't know how to do that since the FSM doesn't mention it. I did notice today though that while I was doing a test drive and holding the throttle partially in during slow acceleration I hit something like a dead spot in the powerband. I saw my AFR go super lean when it happened and then quickly compensate to super rich. Maybe the ECU is just still learning though since I reset it several days ago and keep messing with things.

Do you have NB emulation on your wideband?

Maybe... I have an AEM UEGO wideband and there is an option to make the voltage read anywhere from 0-1 volt so I assume that would emulate a narrowband. Do you suppose it would?

Have you monitored the NB output and input going into the ECU?

No I have not. Does it read 0.5 volt at 14.7 AFR and then go lower when lean and higher when rich? I have no clue so please let me know since this test would be interesting.

Does your CEL blink normal when you jump the pins?

Yes it does. Currently and always I have had a code 42 (vehicle speed sensor). I had some other codes but I got those sorted out about two months ago in an effort to solve this problem, they were ignition related.

Unless you're running really long or really short stacks they won't be a problem

Cool, I didn't think so since I didn't have any problems when I switched to my T3 velocity stacks that I'm still running.

Have you checked ECU codes?

Yes, only a code 42 (vehicle speed sensor). That code has always been there and hasn't caused me any problems in the past.

To me this sounds like a sensor problem. Try disconnecting the MAP, TPS, or both.

I've tried disconnecting the TPS and the MAP but never both. Disconnecting either one makes my car run horribly and both have been swapped out with replacements just to be safe but no improvement was seen.

I later went megasquirt and haven't looked back.

I really feel like this might end up being the case for me at this point. I'm not quite ready to give up on solving this problem before attempting to resolve it with a stand alone EMS though. I'm just concerned that if I don't solve this problem now, I'll later find that it's still there after going stand alone. Then again, that might be alright since the EMS may make it easy to see what the problem really is. Do you think Megasquirt is best? I'm mostly concerned about the ease of installation, so long as I get all the great features of being able to actually see what's going on with all my sensors, which I assume is a standard feature on all stand alone units.

How (exactly) is your MAP sensor getting a signal ? To what and where did
you hook it up ? I've read thru this thread, but I haven't picked that up.

I think it best to answer this question with the following link. http://www.side2sideracing.com/installs/20v-vac/20v-vac-stwbt/index.html

Initially I had it hooked up to #4 with the T connector, exactly like side2side says to do for a Silvertop running a MAP sensor and a Blacktop ECU. That seemed to work okay and it was setup that way before this problem began. In an effort to solve this issue though I have messed with vacuum lines in search of a leak. I also decided to plug those two lines that I was using and switch to #3 instead, there was no noticeable change in the way the engine ran though. I also added the MAP sensor filter/small chamber that Toyota has in-line on a factory Blacktop MAP setup. That seemed to help a little bit with smooth acceleration when using very little throttle input but no improvement to my condition. As far as the actual wiring goes, it was all done by Dr. Tweak several years ago and it was plug and play.

[yoshimitsuspeed]

No I have not, I don't know how to do that since the FSM doesn't mention it. I did notice today though that while I was doing a test drive and holding the throttle partially in during slow acceleration I hit something like a dead spot in the powerband. I saw my AFR go super lean when it happened and then quickly compensate to super rich. Maybe the ECU is just still learning though since I reset it several days ago and keep messing with things.

With a sweep test you will hook up your DMM to the potentiometer circuit of the TPS. Ohms works but I usually prefer to do a volt test with it powered on the car.
In as smooth and linear a fashion as possible sweep the TPS from all the way closed to all the way open then back closed again. It helps to do this a number of times. The ohms or volts should increase/decrease in a very linear rate from closed to open then back to closed. You are looking for a section that jumps around, doesn't read clearly or suddenly jumps to a much different number. This would reflect a bad spot in the potentiometer and if this spot gives weird readings then it could function normally otherwise but any time the throttle is sitting in that spot or travels past it the signal could throw the ECU way off.
It is possible the MAP sensor could do something similar.
On that note I have heard of people having issues with the MAP sensor tied to only one cyl. The harmonics and pressure waves of that cylinder can cause bad readings. It is much better to have all four cylinders go to a vacuum manifold of some sort and then to the stock or other pulse damper and then to the MAP sensor.

Maybe... I have an AEM UEGO wideband and there is an option to make the voltage read anywhere from 0-1 volt so I assume that would emulate a narrowband. Do you suppose it would?

Yes it does. First I would read through the AEM instructions to see what they say. After that google AEM uego simulation or emulation. Lots of info online about it.

No I have not. Does it read 0.5 volt at 14.7 AFR and then go lower when lean and higher when rich? I have no clue so please let me know since this test would be interesting.

[jondee86]

Initially I had it hooked up to #4 with the T connector, exactly like side2side says
to do for a Silvertop running a MAP sensor and a Blacktop ECU. That seemed to
work okay and it was setup that way before this problem began. In an effort to
solve this issue though I have messed with vacuum lines in search of a leak.
I also decided to plug those two lines that I was using and switch to #3 instead,
there was no noticeable change in the way the engine ran though. I also added
the MAP sensor filter/small chamber that Toyota has in-line on a factory Blacktop
MAP setup.

Any line that connects directly to the common vacuum rail should be fine for
taking a MAP signal. However, probably worth checking the in-line "pulsation
damper" that you installed. At least one of the small canisters installed by the
factory is a check valve, and you don't want a check valve in your MAP sensor
line. If you can blow thru it in either direction, no problem. But if you can only
blow thru it in one direction, remove it altogether.

Cheers... jondee86

[The Bloody Panda]

Looks like I got some weird stuff going on with the TPS. One sensor cannot be calibrated anywhere near to FSM spec at 0.8mm throttle opening and the other sensor is all over the place in the sweep test. The sensor that I've been running seems to progress smoothly in resistance until you start pressing the throttle quickly, at which point the resistance will suddenly drop before it begins to rise. This goes along perfectly with what I'm experiencing since if I suddenly press the throttle in half way it's like there isn't any fuel going in. According to my readings on the TPS the ECU would actually subtract fuel rather than add it when I quickly open the throttle part way. It would be just my luck with this car that it would be the TPS since it was the very first thing I ever replaced when I initially tried to fix this problem 1.5 years ago. Back then I didn't have anymore money or time to mess with it so I just gave up and drove the car with the TPS unplugged until about 3 months ago since the issue baffled me. The brand new TPS that I got from Toyota back then makes the engine run even worse than my original TPS. Both have been calibrated using a voltmeter, feeler gauge, and to FSM specs. I've got another TPS on order so I'll update this thread when it gets here.

When I had the MAP hooked up the way that Side2Side recommends (to just ITB #4) it did have a problem running smoothly below 2000 RPM, but now that I've got it hooked up to the vacuum rail it runs completely smooth aside from my other issue. I'm going to check that pulsation damper and make sure I can blow thru it both ways.

[The Bloody Panda]

I briefly removed that MAP filter/canister since it is difficult to blow thru it in one direction but it seems like the engine runs a little bit better with the canister in place. I have it orientated in the exact way that the part number picture shows to hook it up, which is the way that it is easiest for the engine to suck thru. I haven't tried reversing it, I don't see any reason to but the canister can still be sucked thru in the other direction just not as easily. Nevertheless, I'm going to experiment today by hooking up the MAP sensor in the old way that it was before just to see what happens. I figure I've fixed/improved so many other things maybe I should take this one back to the way it was before this whole problem started. I highly doubt it will do anything other than make my engine run worse, but it's only going to cost me time so why not.

I checked the voltage on my Denso O2 sensor and I think it looks good but I'd like to know what you guys think. When my engine is idling the AFR will typically constantly bounce between 14.2 and 15.2 on my AEM wideband. When I checked the narrowband voltage the wideband was actually hanging out around 13.9 at idle this time but the voltage was always 0.88 on the NB at that AFR. When the AFR goes rich the NB voltage would immediately increase to as high as 2.2 if the WB was reading 10. When the AFR goes off the scale lean the NB voltage would drop so low that it would start reading in millivolts. Do these numbers sound correct for what the ECU should be expecting?

I got my new TPS and yet again the resistance for 0.8mm throttle opening cannot be set to FSM specs. I suspect this isn't a problem since 89452-12080 was superseded by 89452-22080, so the FSM specs probably no longer apply to that test because both 89452-22080 sensors that I have read about 20 resistance rather than less than the 0.5 that the FSM is calling for. The original 89452-22080 sensor that I have did throw a CEL though so there is something actually up with it but the TPS that I just got makes my engine run much better and performs flawlessly in a sweep test. However, I'm still going a little lean at half throttle (like 16ish) and I have to wait about two seconds for the AFR to come down into the 14's, but now my AFR isn't jumping all over the place sometimes like it was with my 20 year old TPS. If I hook up VVT though I'm going off the scale lean at half throttle and acceleration disappears, I have to wait like 2 or 3 seconds for the AFR reading to even be on the gauge, so that's like 17's, and then another 3 seconds for the AFR to get down into the 14's. Slowly the ECU is always correcting this closed loop problem but never learning around it entirely. I'm starting to run out of options after 3 months of messing with this and beginning to feel like the only possible solution is stand alone engine management. My VVT cam pulley is ticking nice and loud so I'm thinking about replacing it but I doubt that will fix this issue since everyone says it has no affect on performance.

Anybody got any new ideas or old tests that I'm forgetting about? I'm hoping that the voltage is incorrect on the O2 sensor but I doubt that it is. That NB was responding super quick to changes in exhaust gas, I wonder if the ECU just isn't seeing these changes fast enough anymore. When I go WOT open loop is awesome, AFR instantly drops to 13's and power response is immediate. I haven't tried open loop lately with VVT hooked up but I bet it would be the same. Closed loop just sucks horribly though and is always too slow to respond to AFR changes resulting in frequent leaness, it's like the ECU just never learns the correct thing to do.

[yoshimitsuspeed]

The NB sensor should never go over 1 volt. Were you taking the signal straight from the O2?
On the GZE there is another O2 signal that must be an output from the ECU that tells you NB voltage on a 0-5v output. I got real confused the first time I plugged into that one.

Once you are sure you are reading the actual 0-1v output you could do the O2 sensor tests in the BGB to make sure it is behaving right. I am sure there is a test in the BT FSM but the 88-89 BGB definitely has it and should get you in the ballpark.

The NB covers such a narrow range and the ECU is simple enough that it shouldn't actually look like a steady voltage. It should bounce back and forth from just above .5v to just below .5v. Unfortunately if you have a multimeter with too much smoothing you may not see it bouncing back and forth.

[The Bloody Panda]

Were you taking the signal straight from the O2?

No I was taking it from the backside of the ECU connector.

I just tried taking it from the blue and white wires of the O2 sensor and the results were weird. I'm pretty new to taking all these readings and hardly used a multimeter until I started trying to fix this problem so I'm not sure if I was doing it right or not. I disconnected the O2 sensor from the wiring harness plug after the engine was fully warmed up and idling and got a reading of just over 1 volt. Before that I tried to take the reading from the backside of the O2 sensor connector while it was plugged in but I couldn't get a reading. Is it correct to disconnect the O2 sensor from the wiring harness when checking for a volt measurement? I wasn't expecting to see anything since it wasn't hooked up but it was definitely just over 1 volt on the blue and white wire connector pins. Since the AFR was sitting just a hair below 14.7 at the time I'm wondering if this isn't my problem after all and maybe this O2 sensor reads 0-2 volts. Please let me know the correct way to measure for the 0.5v and I'll try it out since I assume my method is flawed.

BGB? Is that big green book? If so I don't have it. Do you have a good link to a downloadable pdf of it?

I did also do an O2 sensor test just now that was in the 20V FSM but it involved taking voltage at VF1 and E1 of the diagnostic port. I couldn't completely follow the instructions though since I wasn't able to hold RPM at 2500 without a second person. I probably could if I got creative though. The results to that test were not what I expected though since with TE1 and E1 jumped I was getting 5v at VF1 and E1, and without TE1 and E1 jumped I was getting 2.5v at VF1 and E1. I could be interpreting the manual wrong but I'm pretty sure it's saying that anything over 0 volts in either case indicates a failure of the O2 sensor, but then again I wasn't able to hold the RPM at 2500 so my engine was just idling instead.

[yoshimitsuspeed]

http://mr-2.info/
The searchable manual is very helpful.

I assume you are in FI-68 in the BT manual?
I think VF and E1 is the port I mentioned and was thinking about earlier that was 5v instead of the expected 0-1v.

In the 2500 RPM test you are looking for voltage fluctuation. You will need to hold the RPM there as it describes otherwise the test won't be valid. On the AW11 you can rev the motor from the engine bay and watch the tach haha. You could run wires from the diagnostic port to the cabin then hook those to your multimeter so you can watch the output while reving it from inside the cabin. Or try to find a friend.
I could have sworn it said the voltage range you should be looking for but I can't find it. You do want to see a fluctuation though and it should bounce above and below the halfway point which in this case should be 2.5v. In this case you would want to see the voltage go from above 2.5 to below 2.5 8 times or more in 10 seconds. If it fluctuates less than 8 times in 10 sec it should mean something is wrong with the O2 sensor or the ECU.

There are also O2 sensor tests on FI-29

You will need the O2 sensor plugged into the ECU to monitor it properly. Otherwise it should see it's not plugged in, throw a code, and go into a rich safe mode.
The OX terminal in the diagnostic port should also give you a 0-1v reading directly from the O2 sensor.


While you have the BT FSM out I would go through the EFI system check procedures section for the MAP sensor, intake temp sensor, water temp sensor,

[The Bloody Panda]

Just did some more O2 testing. When at idle my O2 sensor voltage bounces anywhere from 0.06v and 0.84v when checking from Ox and E1 of the diagnostic port and hovering around 14.7 AFR, so that seems fine.

When holding the engine RPM at 2500 things don't seem to be testing so good. When I look at Ox and E1 I hang out in between 0.79v and 0.84v while moving between 13.6 and 13.8 AFR, which is maybe alright but I'm not sure. When I look at VF and E1 though there is no fluctuation whatsoever. With TE1 and E1 jumped VF and E1 will just read 4.98v indefinitely. Without TE1 and E1 jumped VF and E1 will read 2.49 indefinitely. I tried two different multimeters but neither would show any movement in the voltage for the VF and E1 tests being conducted at 2500RPM.

I've swapped O2 sensors multiple times and changed the ECU once so far. I haven't tried swapping them out again and doing this test specifically on them but I do know that my AFR has always been lean at part throttle as of late no matter which brand new O2 or ECU that I run with. The FSM only says to replace the O2 and then the ECU, but are there any wires or fuses in particular that could also be likely considered. Unfortunately I still have only a code 42 for my CEL so I don't know what the next best move would be. When I go part throttle quickly while stationary my AFR always instantly adjusts to something good after just an instant of leaness on my AFR gauge, but if I'm traveling it always goes lean for too long. I'd think it had something to do with the code 42 but that code has been there ever since this engine was swapped in and never use to cause any problems with my AFR.

[The Bloody Panda]

FIXXXXXEDD!!!!

I decided to do every test on every page of the fuel injection chapter in the 20V FSM but I didn't get very far before I found my problem. Back in late May I checked STA and E1 for voltage while cranking and came up with 5.5v, it's suppose to be 6v or more but I just attributed the low amount of voltage to my 6lb racing battery and figured that's enough for the ECU and it shouldn't be the problem. Yesterday I checked again and to my surprise every time I cranked the multimeter came up with -0.05v, I checked with another multimeter and same exact reading. So I cut the harness and ran a new STA wire. Now when I crank I get 10.6v at the ECU and my ECU responds quickly to A/F changes while I'm in closed loop, which is completely amazing. I went ahead and plugged VVT back in while I was driving around and sure enough it worked like a champ.

After all the maintenance I've done in the process of finding a solution to this problem the engine now runs significantly better than I've ever felt it. It runs nearly as nice and factory feeling as my manual 2013 Accord does, which I didn't think would be possible but the 6 speed BZR 89661-1A860 Blacktop ECU works beautifully on my lightly modded Silvertop.

I'm not surprised that the STA wire was the problem since I ran a new wire for my starter years ago because the previous owner had had an alarm system installed on the vehicle and the work quality was poor. I was plagued with starter issues from the beginning until I realized it was the wiring and not the starter or my small battery. If I had known about the STA wire back then I could have focused on it as well and ran new wiring for that too since I suspect this issue could be related.

What's your paypal yoshimitsuspeed? Please allow me to send you a small token of my appreciation. You've been the biggest help to me throughout this process. I don't know how much longer it would have taken or if I even could have stayed motivated if it weren't for you. I've been working on this car for 10 years and the build is getting close to being completely sorted. This thing is my precious and your help has been invaluable to me. I can finally move on and continue with the few projects that I have left on this thing. Thank you very much for your help.

Thank you also jondee86, you brought a lot of good knowledge to the table and I appreciate your help too. I wouldn't mind sending you a little gift as well via paypal if you'd kindly hook me up with your address.

Thank you very much to everyone else who contributed. I am grateful.

[jondee86]

Haha Replied on your other thread before I read this one. You did
all the work, and now you are a wizz with a multimeter, so give yourself
a pat on the back. Next time round you will be helping someone else

Cheers... jondee86