scoobian

I'm having some bother with overboost on my 95 WRX
...the FC is set for 1.1bar boost but the boost hits that too quickly (apparently) and overboosts, causing the FC to drop the boost back down to 8psi (o.5bar) or maybe 0.9 bar

...I'm assuming its overboosting?

if anyone has an Apexi and is running around 1bar, or better, could you let me know your duty cycle - obviously this is just for ballparking
currently I'm adjusting mine between 70% and 90% in an effort to get the right 'effect'

thanks

Jolly Green Monster

iTrader: (4)

Varies lots but nearer 48 to 50 depending on turbo and solenoid used.

Simon

silent running

iirc 1 bar is nowhere near that much duty on mine (3 port solenoid & VF35).

scoobian

thanks gents
.....if I run mine at less han 70% it NEVER gets to 1 bar
..more work required!

Jolly Green Monster

iTrader: (4)

what year, turbo and solenoid, how is it plumbed?

scoobian

ah well:
is a 95 WRX...with a legacy engine...RHF5 turbo - but no other markings (however it has a slightly larger 'throat' than normal to the air intake (60mm instead of 56mm - if I remember correctly))
Apexi solenoid
not sure of plumbing

..the car is as bought
Ive attached graphs showing the boost and torque curves etc.

Jolly Green Monster

iTrader: (4)

you need to plumb it..

COM to actuator.. NO to compressor cover..

just need to play with target and duty to get it right..

Simon

scoobian

well,Ive been blatting up and down the local dual carriageway (at legal speeds natually)
results:
0.5bar at 45% = 0.5bar boost....smashing
0.75bar @52% = 0.5 bar boost......hmmmm
0.9bar @ 60% = 0.6 bar boost....oh oh

0.75bar @ 60% = 0.5 bar boost .....gggrrrr
0.75bar @ 70% = 0.6 bar boost
0.75 bar @ 80% = 0.6bar boost

0.9 bar @ 80% = 0.9bar immediately dropping to 0.6bar
0.9bar @ 88% = 1.2bar boost !!!!!!!!!!!!!!!!immediately dropping to 0.6bar

well, thats me fooked!

...could it be the solenoid or actuator is 'slow to operate'?

why drop to this mystical 0.6bar after peaking at higher boost....what controls this lower level?
anyone any ideas?

Jolly Green Monster

iTrader: (4)

difficult to say without looking at it...

only though being have you set the throttle the correct way around?

ETC --> Car Select.. TPS downwards for your year.. although I would be surprized if that was the problem.

Have you tried increasing the Duty at the rpm after the peak? using the variable duty per RPM?

Also try altering the feedback.

Simon

scoobian

...variable duty???? feedback?????
...where do I find that then? - its a new 'un on me!

oh and the throttle is the rifght way (I checked just to be sure)

<ahem>...I posted for helpwith a loose hose from the throttle body - the hose was perished and cracked and snapped off the metal pipework under the inlet maifold (leads off to the left sode of the engine (too dark to make it all out) Ive replaced it with a piece of me windscreen washers tubing (not the rigt bore but it'll do till tomorrow) darent start boosting tho in case it blows it off
...."wonders if this had been the problem with the boost???? - find out tomorrow!"

thanks for your help, so far, chum

Ian

Jolly Green Monster

iTrader: (4)

the pipe sounds very likely to be the problem

Simon

scoobian

well sadly the problem is still there
no change from the dual carriageway blats (suggests the pipe was just about hanging on till I pulled it)

JGM or Andy Forrest: why would the boost rise spectacularly between 3k and 4k, then drop back to the lower figure (usually 0.6 bar unless I have the boost set fo 1bar+ in which case it generally seems to reduce to 0.9bar

...I CAN get 1 bar+, so the turbo CAN produce it....to me it sounds like its being controlled down to the lower levels

note: after hecking last night, the apexi boost solenoid is in place and plumbed correctly, but I also have the std solenoid plumbed/wired in too....is this correct?

thanks for any help, folks!

Jolly Green Monster

iTrader: (4)

no you need just the apexi solenoid plumbed in...

Simon

Jolly Green Monster

iTrader: (4)

is it mapped? not try to get work, just concerned you might break it.

scoobian

the car is mapped - by me and my cousin
...relax, he has a dyno business so knows what he's doing (though not pertickly experienced with Apexi)
anyhoo, after reading a load moe posts last nght, I figured it may be theboosrt solnoid no working crrectly (shut, hence initial high boost, thenopening but not cloing again,hence runnin at a lower boost (dictated by the std actuator setting?)
so...I took a hos off last night and sprayed some carb cleaner in it
...MUCH BETTER!
its still not rght, but seemed to control a lot better

...I'm juyst waiting for the engine to cool off now so I can take all the hoses off and clean them and the solenoid properly!

I'll let you know
(I dont think it neds a full mapping due to the work we already did...it just appears to be the boost I have an issue with)

thanks for help so far!

jaytc2003

iTrader: (1)

could it actually be boost surge your experiencing?
I have a vf10 turbo (small, very small) and to get 1.2 bar I have boost level set to 1.1 bar on the avcr with the duty set to 86.
Also the start duty in 3rd gear is +3 and then 4th and 5th is +2

Gregstar

is the pipe from solenoid to actuator smaller than the other pipes?
If not it should be!!!!!
I found this out on my 200sx and skyline. the hose from solenoid to actuator is 4mm and the rest is 6mm ( think they are the right sizes)

Gregstar

Originally Posted by JSF from SXOC
sure. A turbo works by reacting to load demand, you should always set the boost profile up where the turbo is under the full load it will see, which is in the higher gears of 4th/5th on the 200SX. Once the boost profile is setup you then have a problem in that the turbo wont spool as quickly in the lower gears because the load on the engine isnt as high. To get around this you need to alter the waste gate solonoids duty settings to alter its profile to compensate for this lack of load.

With the AVCR you have trim settings for each gear which enables you to run more solonoid duty in that gear. You also have a setting for feedback speed that the AVCR uses to work out the profile of this trim.

You need to run a larger % increase in the 1st than 2nd, and more in 2nd than 3rd, 4th and 5th dont need any trimming as the engine is fully loaded in those two gears.


It’s probably worth expanding a bit on what the various functions of the AVCR are there for.

The learning functionality isn’t really there to give you the correct boost setting when you are setting this up for the first time; it’s just a by-product of that functionality that you can use it to help with the initial setup. The learning function is there to allow you to have a control system in place that can auto tune the boost profile to take into account variations in turbo dynamics due to temperature and atmospheric changes.

The way the boost is controlled is by altering the duty cycle on the waste gate actuator solenoid, that’s all that controls what boost the turbo produces as far as electronics is concerned.

Now with the AVC-R you have two main settings that are used to achieve the turbo boost level required, the first is the target boost level you are aiming for, the second is the waste gate duty value, both of these are set at the various rpm points you set with your NE point programming. The NE points you choose have an effect on how well you control boost, they can have a major impact on how the initial spool of the turbo is controlled. I’ll come back to that.

When in learning mode the AVCR reads the waste gate duty you set initially in your target setting and outputs the waste gate duty to the solenoid at that value, it then reads the value of the MAP sensor and compares that value to the target boost level you have set. If there is a difference between your target boost level and the real level being achieved it then alters the value in the preset waste gate duty to a different value that it thinks will generate the desired boost next time through that rpm point. It will only do this when you go 80% throttle and above and only in the gear where learning is enabled. It will do this for ever in that gear, slowly honing in on the correct value for waste gate duty that will achieve your target boost level until it no longer needs to alter this value. It will then make small changes should atmospheric conditions change in the future. It needs to be in the right ball park in order to do some fine tuning though, if the waste gate duty is too far away from the correct value, it will never be able to start the auto tune process.

When you are setting the AVC-R up for the first time, the best way to go about this is to switch learning off completely and manually tune the waste gate duty to achieve the boost you want to achieve. This means ideally you need two people, one to drive and one to note the boost being achieved and to alter the waste gate duty to a value that achieves that. Once you have the waste gate duty values correct to provide you with the boost you want, only then should you switch learning on to ensure if conditions change your boost will stabilise around your target values.

In order to stop the AVC-R constantly fighting itself, you should choose just one gear to control the self learn fine tuning, and that gear should be the one that is used the most where the engine is under full load, this is usually 4th gear. The problem with choosing a lower gear for self learn is that the engine isn’t fully loaded, especially in the low-mid range, so with self learning on there you are going to have too much waste gate duty when you give the engine full load in 4th/5th because it will have upped the duty to compensate for a lack of load on the turbo.

Choosing the rpm range you start to control the waste gate solenoid via the AVC-R is critical for a couple of reasons. Firstly the max duty the AVCR can output is 90%, but for initial spool you want the duty to be 100%, so to generate maximum spool you don’t want the AVCR doing anything until the point at which the turbo needs to be controlled externally. This means a larger turbo should have the first NE point higher up the rpm range than a small turbo, so the turbo can spool as quickly as possible without any interference from the AVCR running at its 90% maximum. The second reason is you only have a few rpm points to use and you need to spread these to enable you to have control of the waste gate duty in the most critical areas where the dynamics of the turbo are changing the most. If you are wasting a couple of rpm points early on running these at max to try and generate some spool you are taking these away from the most important area in the mid range where you need fine control of the waste gate solenoid to give you stable boost.

Once you have all these areas setup, you can then move on to looking at how the turbo is performing where the engine isn’t fully loaded in the lower gears. Because the engine revs up faster in 1st and doesn’t have the load on it due to the lower gearing, the turbo will not produce the same boost as it is doing when in a high gear such as 4th. What this means is your engine is producing less power in 1st than it is in 4th. This can be a good thing if you have problems with traction, but if you have good traction or prefer to balance traction on the throttle rather than by limiting power, you can force the turbo to work harder by increasing the waste gate duty on the solenoid. As you go up through the gears you find the amount you have to increase the base duty cycle by decreases because engine load is going up, by the time you are in 4th gear you are in a stable condition so don’t need any increase in base duty for 4th/5th.

Another thing to bear in mind with the in gear % adjust is how you will be driving the car. If you want this setup for road driving where you will be going through the gears normally, covering the whole rpm range you are going to be seeing high engine load around the 4000rpm area. The turbo will respond to this well and will pull good boost without much increase needed in the 2nd/3rd gear adjustments. However, if you are doing a drag type run where you take the engine to the red line, then up shift, you will never see this midrange high load area. What that means is the turbo won’t be able to pull the same boost because the load isn’t as high, so you need to run a higher % increase for 2nd/3rd. Because of this I have 2 sets of values for the % adjust, one set for the road and one set for the drag strip or if I want to nail it from the lights. *cough*

I am not 100% sure how feedback speed is working, but I have found running with it set to 3 in the higher gears and 1 or 2 in the lower 3 gears gives the most stable response with the least oscillation of the boost.

Hopefully this helps explain a bit about how the AVC-R works and helps get your setup in the right ballpark. It’s quite a complicated piece of kit but when you break it down into its component parts its quite simple. It really does help if you have someone experienced to help you set this up though, I am sure there are plenty of AVC-R’s out there that could be optimised to provide much better spool in particular.

scoobian

WOOHOOO!!!!!!!!!!

sorted!!!!!!!!!!!!!!!!!!!!!!!!!!
the boost solenoid was plumbed the wrong way around (as suggested by JGM<above> ) and DESPITE my checking it last night
swapped them over and cleaned the hoses and it now controls boost up to 1.1bar @ 62%cycle!!!!!
I am SOOOOOO happy

well, actually, not THAT happy as I have just been getting theCEL flashing and when I started monitoring KNOCK its MASSIVE on full boost!
...and then the engine died at idle a couple of times at roundabouts
...but Ive found that vacuum hose off the throttle body has blown off
I'll replace it tomorrow......that'll cure all ills wont it?

oh and thans for the essay above - VERY enlightening! <I love all this techy stuff

scoobian

these things are set to try us:
the knock has been steadily getting worse today - I can understand knock, since sorting the boost, but not 'getting worse'??

anyhoo, it wouldnt idle at all and the ignition advance was nowhere near the map setting for speed/load
so I've messed about for a while checking apexi settings and retarding the ignition at boost (where the knock occurs), but STILL no real success

UNTIL......I refitted the inle pipework/induction kit back onto the turbo housing

it must have been knocked while I was frigging about with hoses
(it doesnt attach too well as the turboinlet is larger than standard, so the previous owner simply cut a split into te rubber inlet so it would fit over (I cant find a rubber that WILL fit and replace their bodge, so Ive had to bodge it a little bit further

so anyhoo: the moral of the story is: dont trust anyone elses work and DO trust your instincts!!

<can I just drive my car around now with no further issues...pleez?>

Jolly Green Monster

iTrader: (4)

sweet...

if it was mapped with it the wrong way around, unless you are mapping as you go you should check it..

Simon

Jolly Green Monster

iTrader: (4)

lol.. some cars seem to want attension at least twice a week and others the bonnet rusts shut.. no real reason often.. lol

Simon