PDA

View Full Version : Boost spiking on throttle off



eyeballprawn
14-07-2009, 03:04 PM
I understand about spiking from spooling up too fast, but why would my boost be spiking on throttle off? I have an EBC installed and it will show a steady 10psi but then it spikes as high as 13psi when I let off the throttle at decent revs.

The car is an auto, and it's the same spike in both Tip (if I stay in the same gear) and Auto (if I lift off and let it change up a gear or two). The BOV is stock, could it be that it can't expel the extra boost properly?

The hosing for the EBC solenoid is all as short as can be, however the control unit in the cabin requires a vacuum hose to be run all the way to it. The unit is in the coin pocket that folds down near your right knee.

I've Tee'd off the the line going to the Fuel Pressure Solenoid for the control unit, the manual said to do it there. The unit is also where I'm reading these boost levels from.

Cheers.

peter thomson
14-07-2009, 03:19 PM
Have you checked the BOV is working correctly

eyeballprawn
14-07-2009, 03:26 PM
No I haven't had a chance to have a look at it yet sorry, I only just got back this evening from test driving. However, it does seem to make a release noise at the correct times but I believe the stock BOVs are pretty quiet anyway aren't they? How can I test it, just by removing and blowing air through it?

orionn2o
14-07-2009, 03:37 PM
As you let off the throttle you close the butterfly...

What happens then? Lots of air rushing out of the turbo, big jump in pressure, then a quick fade away and tail off.

Turbo_Steve
14-07-2009, 04:52 PM
Are you sure the release noise isn't the turbos stalling? :D Sounds awesome but not reccomended.

That would explain the pressure increase.

Nick Mann
14-07-2009, 07:46 PM
Where are you monitoring the boost? If you are monitoring pre-butterfly, then the spike is the butterfly closing.

There is no way that I can see that the manifold pressure can increase when the throttle is closed, unless the BOV is not working.

Davezj
14-07-2009, 08:00 PM
agree with nick here!

if boost is being monitored correctly, after the butterfly valve in the throttle body boost should not be able to increase when trottle is released.

bradc
14-07-2009, 09:17 PM
What they all said ^ sounds like normal operation to me. It should only be a very brief spike when you lift off the throttle.

eyeballprawn
14-07-2009, 11:09 PM
Boost is being measured after the throttle body, from the hose that goes to the Fuel Pressure Solenoid. So that's why I can't understand how it can still increase after the throttle is closed.

So if this is normal operation in a turbo car, does anyone care to humour me with some theory as to how it works?

My feeble mind sees it this way:

Letting off the throttle suddenly, the pistons rapidly slow down so the boost pressure in the intake backs up, however normally the BOV should release this. But if the throttle is snapped shut the higher, backed up boost is now stuck in the plenum with nowhere to go and just has to wait until the engine can pass it through.

Does that sound plausible?

Davezj
14-07-2009, 11:16 PM
where exactly is the fuel pressure reg being feed from are you 110% sure it is from thr plenum chamber.

What side of the fuel pressure control solenoid valve are you connected to, as that could be you problem. i would imagine it should be the plenum side of the valve to read manifold pressure correctly and not the messed about with pressure that is controlling the fuel pressure regulator.

do you have one of the translated jap manuals if so then look at vol. 1 page 72, it has the pressure pipe work diagram on it and you can see what i mean, look at diamond shape 3 valve.

eyeballprawn
14-07-2009, 11:44 PM
This is where I've Tee'd into for the EBC control unit/gauge.

Davezj
14-07-2009, 11:52 PM
that looks like the right place to me, as i have my boost gauge on exactly the same point.
However i do not get a boost spike when i lift off the throttle. but my boost is stock about 7-8psi. i only get a spike when i initialy put my foot down at 3000rpm in 3rd, 4th, 5th. then after a few seconds it all settles done and boost is constant.
now we understand you have everything plumed in correctly and you are actully measureing the plenum pressure we can get to the diagnosis. but it is very important that we know what you know. so we are not making assumptions.

i will double check on my car tomorrow and do a 3000rpm in 3rd, 4th in tip mode. and get back to you but i am sure there is no spike on mine when i lift off the throttle.

the only way i can think of that the pressure can increas ing the plenum when the butterfly valve closes is there is some kind of blow back from the cylinders, which would be very bad, valve not sealing properly, etc. but you would expect to see that after every 2 revolutions of the engine when that cylinder fires again and you would see it at idle as well. if explosive gases were being forced back past the valves into the plenunm chamber at idle the boost pressure would surely show up. don't worry i am just thinking out loud.

eyeballprawn
15-07-2009, 12:31 AM
Thanks for that. My EBC has a 'last boost' display that flashes when I let off the throttle. It's this that's showing the spike, however under power it holds nice and steady.

Davezj
15-07-2009, 12:37 AM
does it spike when you initially put you foot down.

When you say last boost, does that mean it is the last maximum boost recoreded, if so then that will be the spike as you put your foot down as it starts to accelerate at about 3000rpm i say 300rpm as this is most likely to give you a boost spike when you put you foot down.

if not then what does it actully record. last boost what is that and how is it beneficial to recored it.

eyeballprawn
15-07-2009, 01:10 AM
No I'm not getting spiking when I initially put my foot down, and if I did I'm able to tune it out by adjusting the START BOOST to spool up slower so that it doesn't overshoot.

From the manual:



LAST BOOST
This mode is used to display the last boost that was recorded every time the accelerator was released for 3 sec.


It's not worded very well, it means the last boost is flashed up for 3 seconds before returning to displaying the normal current reading.

There is also a PEAK boost recorded. If I reset this before a test run in a certain gear, say 3rd in Tiptronic mode, it will correlate with the LAST BOOST.

However.. there is also a GAIN function to help stop the boost falling off at high revs. Unfortunately the controller won't let me reduce this to zero, the lowest I can go is 5%.



GAIN
This mode is used to adjust the boost consistency. (Adjusts the sensitivity of the valve control).
- When boost spike occurs, lower the GAIN control.
- When the boost falls off at high RPM, or the boost increases as RPM increases, raise the GAIN value.


If anything my boost slightly drops through the revs, perhaps 0.5psi, but I would consider that pretty steady.

There's also a LIMITER function that will pull the boost back a certain percentage when exceeded, but since the throttle is shut it has no effect on the spikes I'm seeing.

kc427
15-07-2009, 04:10 AM
I agree with you Dave, once the throttle is closed, and there is a spike, inside the plenum, there must be something to increase the pressure inside the plenum.

I am not sure, but what I can guess is, let's say, when acceleration hard, and suddenly shut the throttle, then, for example, if the BOV does not release the gas rapidly enough back to the intake(pre-turbo pipe), it could cause the turbo stalls. OK, let's assume this is the case, then, if the turbo stalls, ie, the turbine stalls, and the exhaust gas between cyclinder to the turbine has nowhere to go, then, what will happen? I don't know, as I am not a mechanical guy, but I just guess what it might happen.





that looks like the right place to me, as i have my boost gauge on exactly the same point.
However i do not get a boost spike when i lift off the throttle. but my boost is stock about 7-8psi. i only get a spike when i initialy put my foot down at 3000rpm in 3rd, 4th, 5th. then after a few seconds it all settles done and boost is constant.
now we understand you have everything plumed in correctly and you are actully measureing the plenum pressure we can get to the diagnosis. but it is very important that we know what you know. so we are not making assumptions.

i will double check on my car tomorrow and do a 3000rpm in 3rd, 4th in tip mode. and get back to you but i am sure there is no spike on mine when i lift off the throttle.

the only way i can think of that the pressure can increas ing the plenum when the butterfly valve closes is there is some kind of blow back from the cylinders, which would be very bad, valve not sealing properly, etc. but you would expect to see that after every 2 revolutions of the engine when that cylinder fires again and you would see it at idle as well. if explosive gases were being forced back past the valves into the plenunm chamber at idle the boost pressure would surely show up. don't worry i am just thinking out loud.

kc427
15-07-2009, 04:14 AM
OFF TOPIC: Try not to have overshoot when initially step on the gas. I have used EvoScan to log the knock, and I saw there was as much as 3x knock sum counts when overshooting. In this case, the ECU will retard the ignition timing, and I guess you know what will happen if ignition retards. Finally, I had to adjust my EBC to avoid overshooting in order to have good result. :D




No I'm not getting spiking when I initially put my foot down, and if I did I'm able to tune it out by adjusting the START BOOST to spool up slower so that it doesn't overshoot.

From the manual:



It's not worded very well, it means the last boost is flashed up for 3 seconds before returning to displaying the normal current reading.

There is also a PEAK boost recorded. If I reset this before a test run in a certain gear, say 3rd in Tiptronic mode, it will correlate with the LAST BOOST.

However.. there is also a GAIN function to help stop the boost falling off at high revs. Unfortunately the controller won't let me reduce this to zero, the lowest I can go is 5%.



If anything my boost slightly drops through the revs, perhaps 0.5psi, but I would consider that pretty steady.

There's also a LIMITER function that will pull the boost back a certain percentage when exceeded, but since the throttle is shut it has no effect on the spikes I'm seeing.

Davezj
15-07-2009, 07:40 AM
No I'm not getting spiking when I initially put my foot down, and if I did I'm able to tune it out by adjusting the START BOOST to spool up slower so that it doesn't overshoot.

From the manual:



It's not worded very well, it means the last boost is flashed up for 3 seconds before returning to displaying the normal current reading.

There is also a PEAK boost recorded. If I reset this before a test run in a certain gear, say 3rd in Tiptronic mode, it will correlate with the LAST BOOST.

However.. there is also a GAIN function to help stop the boost falling off at high revs. Unfortunately the controller won't let me reduce this to zero, the lowest I can go is 5%.



If anything my boost slightly drops through the revs, perhaps 0.5psi, but I would consider that pretty steady.

There's also a LIMITER function that will pull the boost back a certain percentage when exceeded, but since the throttle is shut it has no effect on the spikes I'm seeing.

i am not famillier with EBC but i would sugest fitting a standard mechanical boost gauge for a short time (tee it in from inside the cabin from the pipe the EBC is connected to) just to check this out in real time as a spike when you take your foot off the accelerator should be visable on this mechanical gauge.
From the wording in the manual it looks like it displaying the peak boost when you take you foot off the accelerator, which occured earlier on in the acceleration run. I know you it is not doing that but that is what it sounds like to is doing from the wording.
glad to see we are getting somewhere now.

orionn2o
15-07-2009, 08:19 AM
Or take a short mobile vid of it happening.
Is it a greddy ebc?
With most ebcs, they ll show the max boost during that particular cycle as suggested above.
Without accurately plotting boost, its difficult to say whats happening.
When i take my foot off the gas, my boost appears to go sky high for a second. However it doesnt, cos it cant! Its just the gauge reading funny.

Ill upload a pic in a min

orionn2o
15-07-2009, 08:38 AM
uploaded/3336/1236934557.jpg

Gowf
15-07-2009, 09:02 AM
OFF TOPIC: Try not to have overshoot when initially step on the gas. I have used EvoScan to log the knock, and I saw there was as much as 3x knock sum counts when overshooting. In this case, the ECU will retard the ignition timing, and I guess you know what will happen if ignition retards.

As far as i am aware, the ecu will only retard after a knock sum of 5 or greater has been logged. I really wouldnt worry too much about a sum of 3. If you use external knock detection, you'l find that the ecu does see knock when its not there, far too sensitive.

As for the boost spike, surely the bov will only open when it sees a pressure differential between pre and post butterfly? It will take a very small bit of time for the pressure in the plenum to stabilise and become vaccuous, so as others have said, it may increase slightly for a very small period of time.

If you've got a stock BOV, they are not the loudest of things, but sound totaly different to a stalling turbo. I personally wouldnt be overly worried about it, but if you are then i suppose the only true way is to have a BOV mounted as close to the throttle butterfly as possible.

Turbo_Steve
15-07-2009, 09:39 AM
Quick question: Where is your EBC Measuring boost from: inlet manifold or inlet pipe?

eyeballprawn
15-07-2009, 10:26 AM
It's measuring from the manifold, see here (http://www.clubvr4.com/forum/showpost.php?p=450334&postcount=11).

I agree that Tee'ing in a normal gauge will give a better idea of what's going on, now to find someone I can borrow one from!

So perhaps the wording of the manual is ambiguous or the EBC is working in an unexpected way. I understood LAST BOOST to show the reading of immediately before I took my foot off the throttle, and PEAK to show the highest boost since last spool.

More testing is required!

wintertidenz
15-07-2009, 11:02 AM
I have found that when planting it and letting off the accelerator at high revs, I got a spike up to 10psi from 7 momentarily. This was measuring from the intake pipe before the throttle body - where the stock boost solenoid plumbs into.

However, if measuring boost pressure from the other pipe on the back of the manifold (the one beside the brake booster) it stayed at 7 and when I released the foot it went back into vacuum.

Note, this is without a boost controller... and I have a VTA BOV. BOV makes the normal sounds, so it sounds like it could just be the quick surge between the butterfly closing and the BOV venting the extra pressure.

Not an expert on this though but that's what it sounds like to me :D

Davezj
15-07-2009, 09:57 PM
yes i would agree with you about the butterfly valve closing and causing a boost spike if your measuring the pressure before the throttle body.

but he is measuring the correct manifold pressure. so i think it is all to do with EBC, and either the set up or it just giving an odd reading.

kc427
16-07-2009, 02:58 AM
oh, maybe I typed it wrongly, actually, I mean knock sum as much as 30+ I had logged, and the ignition was logged retarded a lot when there was overshooting. :shocked:


As far as i am aware, the ecu will only retard after a knock sum of 5 or greater has been logged. I really wouldnt worry too much about a sum of 3. If you use external knock detection, you'l find that the ecu does see knock when its not there, far too sensitive.

As for the boost spike, surely the bov will only open when it sees a pressure differential between pre and post butterfly? It will take a very small bit of time for the pressure in the plenum to stabilise and become vaccuous, so as others have said, it may increase slightly for a very small period of time.

If you've got a stock BOV, they are not the loudest of things, but sound totaly different to a stalling turbo. I personally wouldnt be overly worried about it, but if you are then i suppose the only true way is to have a BOV mounted as close to the throttle butterfly as possible.

VR4WGN
16-07-2009, 09:03 PM
then maby you guys may know my problem too then,sorry if you feel im hijacking guys but my legnum is spiking hard out,just jumps to 20,is it ok for me to restart the process and remove MBC from engine,get boost reading then start again adding boost?,now iv bridged the boostguage off a t form the MBC Exit hose(to intake hose before plenium),and now the guage doesnt go into vaccum out of boost like it usedto when tee'd fro mthe rear of manifold s normal tee positions come from,is there anyhting i may have missed?
~Q~

Davezj
16-07-2009, 09:23 PM
i don't believe the MBC should be plumbed in where you have it connected to. it should be plumbed in, in the line that goes to the stock boost solenoid (has a red band on it). can't remember if it is on the wastegate side or the inlet pipe side. but are loads of threads about where to put it so i would advise you to check up on it.
But the boost gauge can be tee'd in to the smaller of the two line that come from the back of the plenum.

hope this makes sense

wintertidenz
17-07-2009, 03:06 AM
I think the one from the big plastic intake pipe before the TB is what the car is meant to be boosting, and the one from the rear of the manifold is the actual boost. I found that when my boost gauge was tee'd into the one before the TB it wouldn't go into vacuum, but now it's tee'd off the back one it does go into vacuum.

VR4WGN
17-07-2009, 12:21 PM
thnx dave and winter... um ill re plumb them accordingly,as mine is also where you did yours last,so not going into vaccum,,thanx guys
~Q~

eyeballprawn
19-07-2009, 02:25 PM
Does the BOV activate on the gear changes under a WOT run in an auto transmission?

AlanDITD
19-07-2009, 06:07 PM
Does the BOV activate on the gear changes under a WOT run in an auto transmission?

No mate, only when you lift off the throttle

eyeballprawn
20-07-2009, 01:15 AM
I now have confirmation of what is really happening, seems the suggestions that the boost was spiking on the automatic's up changes were correct.. and not spiking on throttle off as I thought. These spikes were so quick that the digital readout couldn't display them in real time and I wasn't seeing them until the LAST BOOST flashed up after throttle off.

So, the Profec's manual is ambiguous and should instead read:



LAST BOOST
This mode is used to display the last peak boost that was recorded since coming into boost every time the accelerator was released for 3 sec.

Davezj
20-07-2009, 01:21 AM
glad you got it sorted.

it is alway good to find out for definate what is going on. glad the suggestions that you got from this site were correct. Some of use do know what we are talking about after all.

eyeballprawn
20-07-2009, 01:32 AM
Yes indeed :) I really appreciated everyone's assistance and couldn't understand why the controller was telling me those readings.

The downside is that I still have the spikes. But I can't really see how to get rid of them with the automatic tranny . Should I even want to get rid of them in an auto or will performance suffer under WOT up changes? Theoretically, if the BOV *did* actually activate, which it can't, would that be beneficial?

Kenneth
20-07-2009, 01:42 AM
It is likely that when the change in RPM alters the engines air consumption rate, the spike occurs.

That being the case, it will always happen.

I wouldn't worry about it.

eyeballprawn
20-07-2009, 05:04 AM
It is likely that when the change in RPM alters the engines air consumption rate, the spike occurs.

That being the case, it will always happen.

I wouldn't worry about it.

Nice concise explanation, that makes sense. All I can do then is to wind the boost down so that it never spikes too high to cause concern.

Kenneth
20-07-2009, 05:13 AM
Log your detonation count and compare to your boost spike.
If it doesn't cause the detonation count to go up, don't worry about it.

eyeballprawn
21-07-2009, 12:10 AM
Shall do, I have just ordered EvoScan.

Out of interest, how fast can the wastegates vent boost compared to a BOV? I'd assume they'd be less efficient?

What I'm getting at is, with a more intelligent EBC could it be possible to reduce this spiking by venting via the wastegates? Or would the lag between the turbos and the EBC reading the pressure at the manifold make this difficult?

Kenneth
21-07-2009, 12:28 AM
Wastegates do not vent pressure, they let the exhaust gas bypass the turbine which results in the turbine slowing down. As a result, the compressor pumps less air and boost drops

I think controlling that spike is going to be difficult no matter how you approach it, mainly because you don't know exactly when the gear is going to change.

Again, if it isn't causing detonation, don't worry about it. All auto VR-4s probably behave this way. (At least ones with the same level of mods)

eyeballprawn
21-07-2009, 12:44 AM
Thanks for clearing up my misconception. I was just thinking ahead to if I ever wanted to do further mods and up the boost and the possibility of hitting fuel cut during these spikes.

Kenneth
21-07-2009, 12:56 AM
If the X axis is seconds, then the spike does not last long enough for the ECU to fuel cut on that alone.

You might find with increased boost that the spike isn't as big.

eyeballprawn
21-07-2009, 09:48 AM
I have taken the previous comments on board, but please just humour me for the sake of discussion & further knowledge.

With a manual car, does the boost drop right off at the gear change as the BOV vents and if so then by how much? Would we expect to see a dip at that point instead of the spike?

What is actually the best possible scenario? Hold boost from where it tapered off to in the last gear, then smoothly ramp up to the maximum desired boost level again without spiking?

Turbo_Steve
21-07-2009, 01:45 PM
The "best" scenario would be for the boost to be held at the target ready for the next gear. This neccesitates keeping the turbo spinning, which requires exhaust gas flow.

So ALG allows air to bleed into the exhaust system (usually via EGR-type pipes) and injects and ignites at the exhaust part of the cycle, resulting in no power, but a hot gas mixtures exploding inside the exhaust manifold and driving the turbo round. A good (read "bloody expensive" ALG system actually maintains a boost target based on the RPM select for the next gear. Most systems are a big cheap and cheerful and just keep the turbo spinning.

ALG will pretty much make your car illegal for road use and emissions as well as noise, puts more torque load on yoru drivetrain (in a manual) and tends to gradually destroy exhaust manifolds and turbos as the temperatures are very difficult to manage, and the forces involved considerable.

It's also virtually pointless on a car with teeny tiny turbos like ours!

eyeballprawn
21-07-2009, 11:36 PM
Sorry Steve, what's ALG? Anti-lag something?

Kenneth
21-07-2009, 11:56 PM
A BOV releases the pressure when the accelerator is lifted. As such it works the same for auto or manual transmissions.

As boost is usually measured from the plenum, the fact that the plenum goes into vacuum is due to the throttle being released (and closing the throttle plate). This would happen with or without a BOV.

The BOV when activated, just lets air continue to flow through the intake system without having to go into the engine.

That being the case, you would expect a pressure dip which looks similar to your log where you lift the throttle.

Neither is good or bad, just different.

AlanDITD
22-07-2009, 12:07 AM
Have i missed the point here, but dont all internal wastegates have a boost spike to some degree. External wastegtes have always been the alternative, and dont produce spikes as the gasses are completely clear from the turbo when they are vented.

Turbo_Steve
22-07-2009, 05:55 PM
ALG=AntiLag, indeed.

eyeballprawn
24-07-2009, 03:53 PM
Log your detonation count and compare to your boost spike.
If it doesn't cause the detonation count to go up, don't worry about it.

I just did a logging run, and in general there is some knock under hard acceleration at high boost, the highest recorded was a knock sum of 14. However, the gear change spikes had only 2 or so. Is that ok?

Davezj
24-07-2009, 11:46 PM
what fuel did you have in the tanks when logging it, as the lower the octane below 100 the more knock you will get under hard acceleration.
i am talking about uk fuel rateings here, don't know if they differ in AU.
if i use 95 RON and try logging knock it can be massive. but if i log with 98 RON or 99 RON in the tank the knock is minimal under hard acceleration. Knock of 2 tops.

eyeballprawn
24-07-2009, 11:56 PM
I'm running 98 RON, perhaps I should try with a tank full from a different service station to eliminate that possibility.

So when people talk about knock count, is it the same thing as knock sum?

Davezj
25-07-2009, 10:06 AM
thats a good point i was monitoring knock sum. which i think is a added count hence the name.

but if i remember rightly it was not a gradual build up and drop off of knock but big spikes.

i think the one you shold be monitoring is knock count, which will give an indication of knock at that particular momenrt in time.

eyeballprawn
05-08-2009, 01:23 AM
During the auto's gear changes Evoscan logs confirm the RPM is dropping while at the same time the boost is spiking.

Apart from this, is there ever a situation where you would actually want or expect the boost to increase while revs went down?

Turbo_Steve
05-08-2009, 07:54 AM
You mean apart from Anti-lag? :)

eyeballprawn
05-08-2009, 12:44 PM
Yes apart from in an anti-lag system. On a normal road car, in everyday driving.