Engine bogging down with any swift throttle change

[JustJeff]

I also put this in my build thread, but that thing has gotten so long and often times mundane that I doubt a lot of people are reading it.

Long story short, engine is up and running but...

But, it still has the same bad bogging and event stalling if the throttle goes up with much of any quickness. Even bumping the RPMS from idle to 2k will cause it to bog and stall. It was doing this before rebuild and I thought I had found the cause in that my charge pipe to cold side coupler was splitting as I was dismantling things for teardown. I thought it had a small tear and my problem was a vacuum and pressure leak. That is not the case.

I'm still bleeding the coolant so I haven't had much time to diagnose anything. What I did notice is that the engine idles better and has less of a problem when vacuum hose that feeds my OEM boost sensor as well as my boost gauge is off the UIM. I use the nipple below the BAC with a tee to feed both OEM boost sensor as well as my aftermarket boost gauge.

When time permits I'll start testing TPS, OEM boost sensor, AFM, etc. Any thoughts on this would be appreciated.

[RETed]

Check TPS first...
Make sure it "reads" linearly from rest (i.e. idle) all the way up to WOT.
Make sure max resistance isn't over 5.5k-ohms.

After checking the TPS, unplug the boost sensor electrical plug and see if this condition changes?

If those two things check out, I would try and eliminate the ignition (spark) side of things, since it's easier at this point...
Make sure all our spark plugs are firing, etc.


-Ted

[Rotary Evolution]

and i wouldn't worry too much until the engine has at least a few hours on it, the idle has been reset, TPS adjusted and base timing set. if it is still doing it do a TPS sweep to check for dead spots and a vacuum leak test. low fuel pressure can also be a problem, as well as poor engine grounding.

[JustJeff]

One inaccuracy. I do not have my OEM pressure/boost sensor routed from the UIM nipple under the BAC. That nipple feeds both OEM BOV and AEM boost gauge. The OEM pressure sensor is routed to/from the cold side turbo splitter, which also is routed to my MBC and then from that to my wastegate.

What I noticed today is that idle jumps up about 500RPM when I pull the hose off the UIM to BOV and the idle evens out. Also if I pull the hose off the OEM pressure sensor or unplug the pressure sensor there is no change in idle.

FSM says to hook up a pressure tester to 100mmHG, turn ignition switch and check voltage. Can I simply blow in the pressure sensor and read the voltmeter?

TPS testing:
Narrow: .8 as I go to WOT it will read once at 1.06 then go into OL and then pick up and have one reading of 6.4 where it will hold

Full: .661 closed and as it goes to WOT it will range up to 3.8, then go into OL and pick back up 4.43 and hold.

[Rotary Evolution]

the compressor outlet doesn't show vacuum, only boost positive pressure and not the same pressure at the manifold leading into the engine so it isn't a good spot for the pressure sensor line. the sensor needs to see engine vacuum and pressure that the engine sees to accurately advance timing.

move the line to either the top or bottom front port just past the throttle body off those 3. the center port iirc goes to a port before the throttle body so isn't a good MAP source.

i highly doubt this is your problem but it is a problem that should still be addressed.

[JustJeff]

the compressor outlet doesn't show vacuum, only boost positive pressure and not the same pressure at the manifold leading into the engine so it isn't a good spot for the pressure sensor line. the sensor needs to see engine vacuum and pressure that the engine sees to accurately advance timing.

move the line to either the top or bottom front port just past the throttle body off those 3. the center port iirc goes to a port before the throttle body so isn't a good MAP source.

i highly doubt this is your problem but it is a problem that should still be addressed.

Yeah, I was thinking about just that while diving home for dinner. I need to get more vacuum hose to make any changes so that will have to wait till tomorrow.

Whats your opinion of using the same nipple that's under the BAC for Pressure Sensor, BOV, and aftermarket boost gauge?

Also I tested my father's S5 N/A TPS resistances out of a growing curiousity... and found the same areas of deadspots on the narrow range where the multitester goes into OL and then picks back up for one single reading and holds. I seem to see this with every TPS I've tested.

[Rotary Evolution]

the lower ports work but the closer the port is to the intake port in the engine the more resonance the signal will get. you can get by in most cases running an unfiltered line to the throttle body ports but the ones off the LIM should have an orifice in each line to each device otherwise they will 'flutter' in the readings.

i dislike teeing off more critical components like the map and pressure sensors. boost gauge and blowoff valve are less significant. albeit less significant to tee off the pressure sensor on a AFM/MAF based system, it is still used for engine timing calculations, so if a line off that tee ever does break it will advance timing to the ambient reading table which could lead to engine damage although i have never had it happen.

[RETed]

Whats your opinion of using the same nipple that's under the BAC for Pressure Sensor, BOV, and aftermarket boost gauge?

#1 priority is the BOV valve.
Always use a dedicated vacuum line for the BOV.
Do NOT "T" this line, if possible.
If the BOV doesn't get a strong enough vacuum to trigger, it fails to open efficiently.
If the BOV doesn't open properly, it'll kill the turbo due to compressor surge.
I tend to use the fitting under the BAC valve for the BOV, since we're only concerned with vacuum.
The BOV opens due to a vacuum signal - i.e. when you lift off the throttle under high RPM's.
Pulsations under other conditions doesn't mean anything to the BOV; it stays closed under anything close to "0" or positive boost pressure.
This fitting also is usually the closest to the BOV in most installs (and with the stock unit), so having a short vacuum hose helps here.

Everything else doesn't really matter.
Unless you're worried about being off a few psi, they are not important.
Stock ECU retards ignition timing under positive boost - it does not affect fuel delivery.
The stock ECU ignition maps are very conservative, does being off a few degrees won't hurt anything.
Most aftermarket boost gauges aren't accurate enough for you to worry about being off a few degrees; even if the gauge is off, it's just a display.

Killing your turbo is a different matter...


-Ted

[JustJeff]

I'll keep all of this under advisement

What about using one of the fittings on the back of the TB for my boost gauge. I'm trying to remember what is back there. Large one feeds my OMP injectors via the splitter. I can't remember where I routed the other two, but I'm going to need to take the UIM off and look over my vacuum routing anyway.

Also where can I buy those orifices/pills for the vacuum lines (sorry I'm still working caffeine into my system so my thinker isn't too good ATM).

[RETed]

No problem using the fittings off the back (or the front) of the UIM.
I don't remember which one was which, but on the front of the UIM, the top and bottom fittings see vacuum - the middle does not.
In fact, if using a boost sensor or boost gauge, it's better to use those, cause they take vacuum off the surge tank; there is less chance of pulsations from the alternating front and rear rotors.

I dunno if Mazda still carries those pills, but I didn't bother with them.
The pill is to keep spikes from getting to the boost sensor, so the stock ECU doesn't trigger the over-boost fuel-cut unnecessarily.
If you have an FCD (or some way to defeat the over-boost fuel-cut), then this is not really necessary.


-Ted

[Rotary Evolution]

i don't recall any of the rear ports having vacuum, i use them for injector air bleed ports and the oil injector air feed port. the front upper and lower and the ones on the passenger side of the LIM split/ACV port are the only ones i use for vacuum+boost.

the orifice you may be able to find in the help section of an auto parts store. it just looks like a vacuum tee with a brass insert in one end(small hole drilled through), your original pressure sensor hose may still have an orifice in it, it is located about 2" upstream from the sensor in the vacuum hose. a vacuum line filter also works but they are a little more difficult to find.

[Pete_89T2]

I dunno if Mazda still carries those pills, but I didn't bother with them.

The pill is to keep spikes from getting to the boost sensor, so the stock ECU doesn't trigger the over-boost fuel-cut unnecessarily.
If you have an FCD (or some way to defeat the over-boost fuel-cut), then this is not really necessary.


-Ted

I disagree with removing the restrictor pill from the MAP sensor line. If you do, a side effect is it will cause the MAP sensor to produce a "noisy" signal even at idle, which can effect your ability to tune the car. I have an Rtek 2.x modified N370 stock ECU, and when I was having Dave B. at Speed1 tune it for me, he initially had problems trying to determine my baseline on the 1st few dyno pulls. He noticed the MAP figures showing up in the Rtek logs were all over the place. Then Dave figured out I was missing the restrictor pill, popped a spare one he had into the MAP sensor line and all was good from there - MAP readings in the log were no longer noisy, and he was able to get a consistent tune from that point on.

BTW, you can usually find generic restrictor pills for 1/8" hose at Advance auto parts or Autozone.

[RETed]

It's odd that it'll affect the RTEK...
The stock ECU uses on open-loop process for idle, so I would think it would ignore the boost sensor signal.

I've gutted the stock ECU a long time ago and have been running a Haltech E8 for years now. :)


-Ted

[Rotary Evolution]

it's less of a problem on stock cars, the Rtek puts more dependence of tuning on the pressure sensor for the lookup tables for load, the stock ECU mainly uses it solely for timing advance.

i have also seen similar results on many cars without a restrictor in the line, even running the pressure sensor as far up stream as possible you sometimes will get resonant spikes in the readings. the Rtek was actually showing a bounce of between 5"Hg and 2psi under WOT full load off the stock port by the ACV on a non turbo car. the readings smoothed out from a 7"Hg bounce to 3-4"Hg bounce by moving it to the throttle body, installing the factory restrictor lowered it to less than 1"Hg fluctuation.

on some cars even with a standalone on turbo engines i have seen 2-3psi fluctuations even locating the source line at the port i mentioned near the throttle body, i always recommend an orifice located near your map sensor to filter the signal. i suspect this also may be due to some lax filtering of the signal on the ECU side of some manufacturers. those were cars that had either no boost controller or manual boost controllers equipped, wastegate cycling was not the cause.. the result was AFRs bouncing from perfect to 1-1.5 points lean.

[JustJeff]

I love all the input!! Feed me Seymour!! Keep all this knowledge coming!!

Here's what I've found thus far. I change the source for my pressure sensor to the UIM front top port and that did make a significant difference. But I still had a low RPM at idle and if I fluttered the RPM it would still hesitate and bog down.

Throttle cable was part of the problem. It had a lot of slack on it at the back of the TB. I tightened that up and that combined with changing the source for the pressure sensor has made a world of difference. It lops a little bit at idle but idles between 500-900ish idle. I'd raise RPMs up to 2-3k and let them drop back down and idle point would be a little different each time.

It still does a very small hesitation when I first flutter the RPM. Associated with that little hesitation I can hear something out of the ordinary coming from the engine. Sounds like something opening or closing, possibly a vac/pressure source...or maybe the throttle butterflies. It is always associated with the hesitation and only happens if I flutter the throttle. I was working alone so I couldn't stick my head in the engine compartment while someone else fluttered the throttle, but tomorrow I can get a helper.

I did not check timing today because I could not remember which timing marks were which and didn't have a spare main pulley to compare. I had mine powdercoated and then repainted the marks myself. IIRC first one is red/trailing, second one is yellow/leading?

One other thing very strange which may or may not be related to the problems. I have a huge amount of exhaust spray of soot and moisture. It was doing this after the initial rebuild and is still doing it. When I play with the RPMs and get them bouncing up and down, not higher than say 3k I have about a 4-5 spray.
http://i191.photobucket.com/albums/z259/justjeff_photos/Exhaust%20spray/ExhaustSpray.jpg (http://s191.photobucket.com/user/justjeff_photos/media/Exhaust%20spray/ExhaustSpray.jpg.html)
http://i191.photobucket.com/albums/z259/justjeff_photos/Exhaust%20spray/Consitencyofspray.jpg (http://s191.photobucket.com/user/justjeff_photos/media/Exhaust%20spray/Consitencyofspray.jpg.html)

I didn't get a close enough pic, but it's moisture with soot. I tasted the moisture and it's nothing but water...no coolant. Where is the moisture coming from? Is all of that from condensation in the exhaust?

[Rotary Evolution]

condensation is normal so i wouldn't exactly say that the spray from the exhaust is a problem but just that the engine isn't warm yet.

standing at the front of the car looking at the engine the mark on the driver side of the pulley is the leading yellow mark, the one to the left is the trailing red mark. the engine MUST be idling below 1k RPMs to set timing otherwise it will advance and not be set properly.

warm up the engine and idle it down to 1k or less(you will hear the engine begin to labor and smooth out when it is in idle mode which turns the engine under base timing until it jumps up off idle), jumper the initial set coupler(not really necessary but it will disable the BAC valve which may hold it above 1k). hook up the timing light to your lead #1 rotor plug, adjust the CAS so that the closest yellow mark lines up with the pointer and lock the CAS. move the pickup to trailing #1 plug wire and it should line up with the far red mark.

to test the TPS, hook a DVOM black lead to battery negative, positive red lead to a small pin jammed into the green wire off the TPS. adjust the screw so that the TPS reads 1 volt with the engine warm and idling at approximately 750-800RPMs.

do a sweep of the TPS with the key on/engine off/engine warm, slowly open the throttle to 1/3 and it should read approximately 4.6v and not have any jumps or drop offs in voltage while doing the slow sweep.

if the idle is being problematic and won't idle down to under 1,000RPMs with the engine fully warm then the thermowax is either stuck open due to an air pocket in the cooling system, is stuck due to age, or the idle air bleed screw needs to be turned in. do this first before anything else, sorry for putting it last.

TPS issues/adjustment and timing are the usual causes for off idle stumbles but it could go deeper to a fuel system problem or a vacuum leak.

if the noise you are hearing when romping on the throttle is a pinging sound, like a ballpeen hammer rapping on a sheet of thick metal rapidly then the timing is too far advanced. your timing pulley on the engine may also not match the hub on your engine if it was ever mixed up before during a rebuild. this is usually distinct in that the engine usually kicks back randomly when trying to start the engine. if this is the case then fix it now or the engine will be short lived as soon as it sees boost.

severe advance knock sounds like you're hammering on the engine with a sledge hammer once and then fades rapidly as soon as engine RPMs come up, when you stomp the pedal off idle.

[RETed]

on some cars even with a standalone on turbo engines i have seen 2-3psi fluctuations even locating the source line at the port i mentioned near the throttle body,

I've never had a problem with this, especially with Haltech systems.
I blame this on a badly tuned fuel map (at idle).


-Ted

[RETed]

I didn't get a close enough pic, but it's moisture with soot. I tasted the moisture and it's nothing but water...no coolant. Where is the moisture coming from? Is all of that from condensation in the exhaust?

Wow, you're crazy!
Did you type "tasted"???

Don't worry about it until the engine is properly broken in.
Low compression + blow-by tends to spit all kinds of weird stuff out the exhaust.
Couple this with the car sitting around for a while, stuff gets built up in the exhaust...


-Ted

[Pete_89T2]

It's odd that it'll affect the RTEK...
The stock ECU uses on open-loop process for idle, so I would think it would ignore the boost sensor signal.

-Ted

That's true, what I meant to say was that the MAP sensor readings we saw in the Rtek logs were noisy even at idle; idle itself was smooth and steady. When doing the baseline dyno pulls, we couldn't get consistent HP & torque numbers from pull to pull, so after checking the logs, it became obvious that the noisy MAP sensor readings was the cause, and the restrictor pill solved that problem.

[Rotary Evolution]

I've never had a problem with this, especially with Haltech systems.
I blame this on a badly tuned fuel map (at idle).


-Ted

how does a poor fuelling map translate to 2-3 psi fluctuations at 10-15psi @ 11:1 AFRs? note i mentioned it wasn't a common problem but something i have noticed before on a few cars. if the cars were misfiring due to overly rich conditions in the sub 10's i would tend to agree.

[JustJeff]

condensation is normal so i wouldn't exactly say that the spray from the exhaust is a problem but just that the engine isn't warm yet.

standing at the front of the car looking at the engine the mark on the driver side of the pulley is the leading yellow mark, the one to the left is the trailing red mark. the engine MUST be idling below 1k RPMs to set timing otherwise it will advance and not be set properly.

warm up the engine and idle it down to 1k or less(you will hear the engine begin to labor and smooth out when it is in idle mode which turns the engine under base timing until it jumps up off idle), jumper the initial set coupler(not really necessary but it will disable the BAC valve which may hold it above 1k). hook up the timing light to your lead #1 rotor plug, adjust the CAS so that the closest yellow mark lines up with the pointer and lock the CAS. move the pickup to trailing #1 plug wire and it should line up with the far red mark.

to test the TPS, hook a DVOM black lead to battery negative, positive red lead to a small pin jammed into the green wire off the TPS. adjust the screw so that the TPS reads 1 volt with the engine warm and idling at approximately 750-800RPMs.

do a sweep of the TPS with the key on/engine off/engine warm, slowly open the throttle to 1/3 and it should read approximately 4.6v and not have any jumps or drop offs in voltage while doing the slow sweep.

if the idle is being problematic and won't idle down to under 1,000RPMs with the engine fully warm then the thermowax is either stuck open due to an air pocket in the cooling system, is stuck due to age, or the idle air bleed screw needs to be turned in. do this first before anything else, sorry for putting it last.

TPS issues/adjustment and timing are the usual causes for off idle stumbles but it could go deeper to a fuel system problem or a vacuum leak.

if the noise you are hearing when romping on the throttle is a pinging sound, like a ballpeen hammer rapping on a sheet of thick metal rapidly then the timing is too far advanced. your timing pulley on the engine may also not match the hub on your engine if it was ever mixed up before during a rebuild. this is usually distinct in that the engine usually kicks back randomly when trying to start the engine. if this is the case then fix it now or the engine will be short lived as soon as it sees boost.

severe advance knock sounds like you're hammering on the engine with a sledge hammer once and then fades rapidly as soon as engine RPMs come up, when you stomp the pedal off idle.

I've never tested TPS using voltage, I've always done a resistance test. I'll try it your way. Though I didn't understand "with the key on/engine off/engine warm"

A couple things to note, I have a JDM engine so I'm missing a few of the standard USDM parts. I have no thermowax (that is the part sitting on the back side of the TB with 2 coolant ports, correct?) Air bleed screw is on the BAC correct? JDM BACs do not have that screw. What I found yesterday was changing the tension on the throttle cable changed idle RPM. Not sure this is the best methodology, but it worked yesterday.

Pulleys should match as I've only ever worked on S5. So unless a previous owner mucked it up all my pulleys should match.

The noise I hear and bogging/hesitation is when I lightly flutter the throttle. For the most part I can smoothly rise and fall the RPM. It's that swift little start to the RPM going up when it behaves oddly.

Wow, you're crazy!
Did you type "tasted"???

-Ted

Haha, I did and I am. Now mind you I did not put a glass under my tailpipe and take a swig. I simply dabbed my finger in the moisture to see if it tasted like anything other than condensation (coolant, oil, gas)

[Rotary Evolution]

yea, the thermowax has a yellowed plastic port with 2 vacuum ports on it although the JDM engines didn't have that piece on it. the thermowax is what connects to the coolant port on the bottom of the throttle body to the rear iron so if it is missing then it must be deleted.

the JDM BAC has no idle air bleed in the BAC so the idle has to be hard set with the throttle stop screw on top of the throttle body facing the passenger side of the car with an 8mm locknut on it.

i prefer the voltage method over resistance because with older wiring you may get resistance within specification but once the wires have a load on them problems arise which is easier to spot with voltage under load than with a resistance method. an auto ranging DVOM will be a little tricky since it usually will auto range once you hit between 2 and 3 volts, causing the DVOM to lose the reading for a split second. you can choose your own range though so it will not auto adjust.

[JustJeff]

yea, the thermowax has a yellowed plastic port with 2 vacuum ports on it although the JDM engines didn't have that piece on it. the thermowax is what connects to the coolant port on the bottom of the throttle body to the rear iron so if it is missing then it must be deleted.

the JDM BAC has no idle air bleed in the BAC so the idle has to be hard set with the throttle stop screw on top of the throttle body facing the passenger side of the car with an 8mm locknut on it.

i prefer the voltage method over resistance because with older wiring you may get resistance within specification but once the wires have a load on them problems arise which is easier to spot with voltage under load than with a resistance method. an auto ranging DVOM will be a little tricky since it usually will auto range once you hit between 2 and 3 volts, causing the DVOM to lose the reading for a split second. you can choose your own range though so it will not auto adjust.

Thanks for the explanation on thermowax, I do have the coolant port on the bottom and am using it.

From your description it sounds more reliable for me to be setting idle with hard set than by tensioning my throttle cable.

And yes checking voltage does seem more reliable, I assume that using multimeter is the same idea behind using one of the test light tools..only you are looking at actual numbers

I don't know that I'm going to get much work done on the car. Drinking 2 bottles of wine seemed like a splended idea last night, but I'm not very motivated to get much done today.

[vrracing]

My JDM S5 Tii has a thermowax. Here's a picture of it.

http://i366.photobucket.com/albums/oo104/vrracing/RX7Forum/84fafb56.jpg (http://s366.photobucket.com/user/vrracing/media/RX7Forum/84fafb56.jpg.html)

and a close up

http://i366.photobucket.com/albums/oo104/vrracing/RX7Forum/74cb65ea.jpg (http://s366.photobucket.com/user/vrracing/media/RX7Forum/74cb65ea.jpg.html)

[Rotary Evolution]

meaning it has no double throttle vacuum port(the yellow 2 port plastic bit). all the throttle bodies have a thermowax.

[vrracing]

Yes, Jeff said he had no thermowax because he had a JDM. My pictures were intended to show him how to recognize it. If he compared the pictures to his setup and didnt really have it, then he'd know someone molested the engine before he got it.

[JustJeff]

Yeah, I always thought the thermowax was only the plastic fitting with the two ports. Thanks for the explanation.

By the way, that is a gorgeous and clean TB. I have a spare one I'm going to tear apart, clean, etc. Do you happen to have a write-up on cleaning and reassemble?

[vrracing]

Anal-retentiveness FOR THE WIN!

It's just cleaned with Oil Eater. It doesnt cause the white scaling like the Castrol SuperClean does and it doesnt burn my eyes when I spray it. After I cleaned it I lubed all the moving parts with Royal Purple Maxfilm.

There are write-ups on the other forum covering powder-coating the TB. I followed them when I cleaned and reset everything. btw, I set the thermowax on the counter using a heat gun to move it thru its paces.

If I take it off again I'll probably paint it with Duplicolor aluminum wheel paint. I painted the XKE's Stromberg carbs with it and they are beautiful. Or maybe not. The TMIC covers it up. The rotor housings are diamond graphite and the irons red but no one can see em!

[JustJeff]

Nice job on the obsessiveness, it looks clean enough to eat off of

I'm planning on getting a few items powdercoated, my spare TB, alt pulley and any other small items which strike me as looking good in black

[JustJeff]

Here's what I've got so far:

I've got timing close, but trailing timing does not line up. Leading is perfect on yellow (white in my case because I powdercoated and repainted the marks) mark, but trailing is off by a bit. It's right around 1/3 of the way between red and yellow.

I tested TPS a few different ways. I tested straight resistance per FSM and did some deadzone. But I see deadzones in every TPS I test in the same general area. I got advice on the other forum about always testing TPS in continuity mode because sometimes autoranging will "timeout" and show OL. Testing in continuity shows smooth figures with no deadzones.

I also tested with voltage as suggested here and that comes up different. I get it set to 1v, engine warmed up, pos lead back probing greed/red wire, neg lead grounded to battery. Idle might be a little higher than 750-800. It was at 900 when I last had it idling and warmed up. I started up the engine to check idle and turn it down a bit and it was perfect between 750-800. I start moving the idle and it reaches about 3.6v and OL and I haven't gotten to 1/3 of the movement. I made a video on my smartphone. I'll upload it once I get home today/tonight.

[JustJeff]

I got home last night, got into relax mode, and forgot to upload the video. I did do some reading and found opposing info about timing and turbo rotaries. I read that trailing doesn't matter nearly as much as leading. That some don't even check trailing after setting leading. But then I also read that timing on turbos in general is much more important on turbo and than n/a. I found that in rotary tech of the other forum so it might have been more geared towards stand alone controls.