Ignition Advance

[mattatk]

G'day again,

Noticed today on my GoLINK app that my ignition advance at idle is hovering at 5deg, but occasionally cycles between 3 & 5deg, whereas at 3000+ RPM it is steady at 35deg. Is it normal to fluctuate at idle? Having said that my idle wanders slightly @ 685-705 RPM. I feel that even though that those idle figures are normal, you can feel a slight hunt or "burble" in the idle. The car is warmed up

My Maintenance manual only has the 2.5lt engine parameters which state ignition advance @ warm idle should be 5 +/-1.25deg, so these figures are within limits, but should they fluctuate?

Thanks in advance,

Matt

[mattatk]

Further, my DME diagnosis manual (factory) states that the ignition advance @ warm idle should be 5.3 +/- 0.5 degrees crank.

The parameters in the manual are for Bosch DME 5.2, my 2001 986 S has a Bosch DME 7.2, should the numbers be the same? If not does anyone have a listing of the warm idle figures for my car please?

I've changed my spark plugs, cleaned the MAF sensor, cleaned the throttle body & changed the air filter. The other parameters look OK (although idle rpm fluctuates between 675 & 710 - yes within limits but still hunts a bit) except my camshaft deviation is out of limits at a steady 8 degrees (can only monitor one side as it is ROW car).

The car feels a bit down on power/acceleration, as posted separately (I do accel runs with a iPhone dyno app which may not be that accurate but good for picking up trends).

Anyone got any ideas for what could be causing deviations in ignition advance/timing? or the loss of power on accel please?

Cheers,

Matt

[wvicary]

The DME uses ignition timing to control idle speed in a narrow range, if it can't control the idle it then adjust using the throttle body or idle speed motor. So yes it's normal for the timing to be changing at idle but it can also point to a problem. Your idle speed should be steady within approx 25-50 RPM.

Does your camshaft readings ever change? It being fixed at 8 degrees sounds odd to me but I'm not sure on ROW vehicles.

[logray]

The parameters in the manual are for Bosch DME 5.2, my 2001 986 S has a Bosch DME 7.2, should the numbers be the same? If not does anyone have a listing of the warm idle figures for my car please?

Here you go. 986 DME setpoints attached. Yes the numbers should be the same between ME5.2.2 and ME7.2.

Also I've been told from a very reliable source that cam deviation is OK out to +/- 6 degrees, even though listed as 4 degrees pos or neg.

I've changed my spark plugs, cleaned the MAF sensor, cleaned the throttle body & changed the air filter. The other parameters look OK (although idle rpm fluctuates between 675 & 710 - yes within limits but still hunts a bit) except my camshaft deviation is out of limits at a steady 8 degrees (can only monitor one side as it is ROW car).

You should address the 8 degrees deviation because that is 2 degrees beyond (advanced) than it is supposed to be. My car has the most power throuhgout the range when cam timing is 0 degrees. The cams should be checked for proper timing and re-timed. You may also be having a problem with an actuator. With a warm engine, you should log your "actual angles parameters" at idle, and then again at 3000 rpm. And then post the results back here.

I've also wondered on a RoW car even though they may only have one cam position sensor, if one could move that sensor to the other bank and then gather more readings. I suppose it would be possible even to purhcase another CPS, install it, and reflash the DME to USA.

The car feels a bit down on power/acceleration, as posted separately (I do accel runs with a iPhone dyno app which may not be that accurate but good for picking up trends).

It could be a problem with the variocam system, and/orthe car is too far out of timing which will both affect performance, You might also check that the throttle body is clean and there are no intake leaks. If your car has one, you might also want to clean the idle air control valve.

Edited September 30, 2011 by logray

[mattatk]

Thanks Logray for the reply yet again! Yep, my manual's figures are the same.

I'll investigate the reflash idea, I've noticed that the ROW cars are pretty scant on data gathering. My understanding is that the ignition advance is adjusted by the DME to optimise the ignition & obtain info from the knock sensors (obviously to prevent detonation). Could it be possible that the knock sensors are driving the ignition advance back & forth? (correctly or erroneously). What other sensors feed the ignition advance solution?

My laptop is down at the moment, so I'll post the "actual angles parameters" early next week.

This link;

http://986forum.com/...timing-m96.html

details a simple retiming procedure (appears pretty similar to the workshop manual on initial inspection). I am guessing this is what you mean. I've carried out the procedure as far as checking the exhaust cam's notches & they do appear to be slightly off time, but I haven't had the time (courage?) yet to adjust it. If I loosen the sprocket bolts, could there be any tension in the system that could cause the timing to jump?

Thanks again for your tireless efforts, it is wonderful to be part of this excellent website.

Matt

Edited October 1, 2011 by mattatk

[logray]

...I'll investigate the reflash idea, I've noticed that the ROW cars are pretty scant on data gathering.

Yes the suggestion is temporary, since your car doesn't have a secondary air system or extra O2 sensors, and other codes might have to be set in the DME this might purly be to gather more data, but I doubt you would want to run a RoW car like this , might have trouble with inspections as well... so this is a pretty odd ball idea I had.

My understanding is that the ignition advance is adjusted by the DME to optimise the ignition & obtain info from the knock sensors (obviously to prevent detonation). Could it be possible that the knock sensors are driving the ignition advance back & forth? (correctly or erroneously). What other sensors feed the ignition advance solution?

Certainly knock sensors could play a role, but if they are out of spec then I believe they will throw a CEL. IF I knew the answer to your question I probably would work for Porsche, but my guess is things like temperature, lambda, air flow, etc. etc. etc. all affect the timing adjustments, the exact specifications are known probably to those few that have reverse engineered/decompiled the DME's coding and can reflash it.

http://986forum.com/...timing-m96.html

details a simple retiming procedure (appears pretty similar to the workshop manual on initial inspection). I am guessing this is what you mean. I've carried out the procedure as far as checking the exhaust cam's notches & they do appear to be slightly off time, but I haven't had the time (courage?) yet to adjust it. If I loosen the sprocket bolts, could there be any tension in the system that could cause the timing to jump?

Yes this a wonderful post that I always recommend to people. As for loosening those sproket bolts, you are 100% ok to do this, so long as you don't drop a tool or bolt down in the engine and then be sure to retorque them to the proper value with loctite when you are done. I also recommend getting a genuine timing tool, otherwise you are just going to be guessing (as the original author was doing). Also I might caution you that even though you may correct the timing, there could still be another issue present, for example excessively worn cam pads, stretched chains, or actuator issues. So I would consider re-timing a temporary fix if that should restore your power and correct whatever issues you were having, eventually timing chains/actuators/variocam pads do need to be replaced.

Edited October 1, 2011 by logray

[mattatk]

Finally got laptop working again!

Warm idle is OK but fluctuating 683-715 rpm.

Ignition timing/advance fluctuates 3.75 - 7 degrees crank (mostly sits around 5.25 degrees)

Camshaft 1 deviation steady @ -7.92 but as car warmed up it showed -7.56 then later -7.45 degrees (is it actually changing or is it possible error on data/sensor?)

ACTUAL ANGLE -0.5 degrees (with occasional deviations down to -0.33) & then up to 23.3 degrees @ 3100 rpm (couldn't hold it at 3000 rpm)

All other parameters fine except:

Rough running 11.4 (too high, 6.3+/-1.5)

Rough running on individual cylinders holding @ 0 with occasional spikes to -0.5

FRA bank 1 1.12 steady (too high, limit 0.96 +/- 0.03)

FRA bank 2 1.09 steady (ditto)

RKAT range 1 bank 1 -2.53

RKAT range 1 bank 2 -2.2 (are these RKATs the TRA range 1 in the workshop manual??)

My next step is to do the timing adjustment as above & re-record the figures.

I can't get my head around the FRA & TRA or RKAT figures, the RKAT or TRA figures at -2.2 are WAY below 0+/-0.06, does that mean she's trying to add a lot of fuel or that she's to rich & trying to lean it out?!)

Cheers again,

Matt

Edited October 3, 2011 by mattatk

[logray]

Warm idle is OK but fluctuating 683-715 rpm.

Did you try cleaning your throttle body and idle air control valve? That's pretty normal though if it doesn't hunt continuously.

Ignition timing/advance fluctuates 3.75 - 7 degrees crank (mostly sits around 5.25 degrees)

This sounds within limits to me.

Camshaft 1 deviation steady @ -7.92 but as car warmed up it showed -7.56 then later -7.45 degrees (is it actually changing or is it possible error on data/sensor?)

It's too bad Porsche didn't install another CPS on ROW cars. Anyways, -8 degrees is certainly out of spec. The WSM says +/-4 degrees from 0 is acceptable, however I've been told that 6 degrees either way is acceptable (albeit still out of time). When the cams in each bank are timed at the same, this will produce the best results for the engine. At least on my car it ran the best with 0 degrees, but I have experimented with 6 degrees advanced and retarded.

The moving part of your data is what disturbs me a little. Perhaps your sensor is worn. Perhaps it is just Durametric's ability to read fine grain. However, nearly 0.5 degrees at the crank change from cold engine to warm means the cam timing is changing by 0.25 degrees while running the engine. The cam deviation is not supposed to change. If it's not the sensor or Durametric being too sensitive, this could be attributable to timing chain stretch, variocam pad wear, or actuator issue (or worse such as sprocket slipping or ims bearing) and perhaps a reason to open the cam covers. I would get a second opinion on this.

When you put the camshaft deviation on a graph, does it move constantly, or do these fluctuations only happen slowly as the car warms up.

There is information out there that a fluctuating cam deviation could indicate the IMS tube is wobbling due to a failing IMS bearing, however

It's too bad we can't see bank 2... although at least on a 996 it isn't too hard to remove the cam bearing cap for the exhaust cam on bank 2 to see the timing mark and get a timing tool in there.

ACTUAL ANGLE -0.5 degrees (with occasional deviations down to -0.33) & then up to 23.3 degrees @ 3100 rpm (couldn't hold it at 3000 rpm)

This indicates normally functioning variocam actuator on bank 1.

All other parameters fine except:

Rough running 11.4 (too high, 6.3+/-1.5)

Rough running on individual cylinders holding @ 0 with occasional spikes to -0.5

Might indicate/correlate with the cam timing deviation.

FRA bank 1 1.12 steady (too high, limit 0.96 +/- 0.03)

FRA bank 2 1.09 steady (ditto)

RKAT range 1 bank 1 -2.53

RKAT range 1 bank 2 -2.2 (are these RKATs the TRA range 1 in the workshop manual??)

I believe RKAT (short term adaption at idle) is TRA. FRA long term adaption a little high indicates lean if I remember right, the DME is trying to add fuel. and your negative TRA at idle means rich since it's trying to remove fuel. Also have you checked for intake vacuum leaks and cleaned your MAF?

Edited February 16, 2012 by logray

[mattatk]

Yes throttle body cleaned about month ago when replacing faulty AOS, no idle control valve as it's a egas model. MAF cleaned as per Pelican Parts tasks.

Idle does hunt continuously (though within limits)

Ignition advance mainly stays in limits but the way it hunts & sporadically drops outside of 5.3+/-0.5 makes me wonder if it is causing the idle fluctuations.

On a graph the cam deviation looks rock steady, but when I re-check the figures (re load the actual value screens on durametric) the cam dev has dropped slightly, towards the limit +/- 4 deg. I don't actually see the figure drop on the graph, that's why I thought it might be a sensor or software issue.

Do you know limits for actual camshaft angles at idle & 3000 rpm?

I'll get onto the timing & check the vacuum & post any findings in case someone else could benefit from this.

Thanks again SO much, I better donate again!!!!

Matt

Edited October 4, 2011 by mattatk

[logray]

Do you know limits for actual camshaft angles at idle & 3000 rpm?

Actual angles should advance by about 25 degrees depending on RPM/temp. (we don't know bank 2 because you don't have a CPS there).

Your cam deviation is 23.3 degrees advanced so what you saw is pretty good, just about on the mark and probably not cause for concern..

Here is a good read about variocam in your car:

http://forums.rennli...s-a-primer.html

How many miles and what year of car again, not sure if we got that yet?

Fix the cam timing and then check all the values again. Both banks should be at 0 degrees. Again I do recommend a genuine cam timing tool instead of just eyeballing it.

Edited October 4, 2011 by logray

[mattatk]

It's a 2001 Boxster S with 110000km. I'll chase down a timing tool & post the findings eventually.

Thanks again & thanks for the good variocam link.

Cheers

Matt

[wvicary]

Before tearing into the engine I would first cover the basic item and be sure there is no vacuum leaks and the MAF sensor is reading within spec.

[mattatk]

Thanks for the advice, I've just rechecked the MAF, it seems OK (had a clean last month) & shows 15kg/hr at idle & 50kg/hr at 3000 rpm.

So far, I've found a crack in the small air hose from resonator cross tube in intake (very narrow hard plastic line that runs to the left front of engine.

[mattatk]

OK, borrowed timing tool, got parts & got stuck into the cam timing as per the link above.

All went well except it's pretty tight with engine in car! The only problems I came across were;

1. green camshaft plug.

While stabbing the plug to pull it off in a tight spot, I popped the plug into the engine!!! With some strong narrow wire I managed to hook it back up to the entry point & pull it out. Was a tense 20 minute diversion!

2. new torx head bolts for the scavenge pumps.

At the rear right of the engine, I couldn't get the driver with a torx adaptor onto the lower 2 bolts due to clearance from a structural member. I bought 2 normal bolts, the same spec as the old 10mm hex head bolts, & used loctite.

Those scavenge pumps were pretty tight & messy (pre lubricated with oil) to get back into the engine.

Turned car over & it idled nicely. BUT....cam deviation bank 1 now shows -9 degrees!! Worse than before the op, -8 degrees. I've reviewed the procedure & all done correcty. Does that mean my sensor is out? When I get back from work i'm going to put the underbody supports back on & take her for a test drive.

[logray]

Glad you got that sorted. Should be fine with the loctite.

My guess is you rotated the crankshaft the wrong way while the sprocket bolts were loose (it actually works opposite from what is intuitive). In other words, rotate crank clockwise retards cam timing while counter-clockwise advances cam timing. So you need to rotate counter-clockwise appx 1 inch (guess) with the cam sprocket loosened.

Also did you fix that tube leak?

Edited October 7, 2011 by logray

[mattatk]

Yep fixed the air leak.

With regards to the cam timing, that was my initial thought too. Although, I followed the procedure to the letter. I locked TDC compression & checked the timing bank 1, which was out a small amount, so I rotated the crank forward another 1.5 turns & watched for the timing to line up, loosened the cam sprocket bolts (de coupling) & then adjusted the crank back to TDC compression (needed to rotate counterclockwise a fraction, approx 8mm at the circumference of the wheel). Same procedure for bank 2 with a similar adjustment.

The Durametric shows an increase in camshaft deviation by a degree (from 8 to 9 degrees deviation), so if it & the sensor are correct, I need to reverse my adjustment by at least 8mm clockwise to go back to where I started & then approx 64mm (8 degrees x 8mm) further to bring the deviation back to zero. The problem then is that the camshaft tool won't fit in the slots as they will be off the vertical ????

Reading the DIY guide, I want the marks on the camshaft ends to line up vertical (ie. hold the cam timing tool) & have the crank at TDC (compression for bank 1 & exhaust for bank 2)....Is this incorrect, as that is where I'm at?

[logray]

It sounds like your procedure is sound so you might be dealing with other variables at this point if the cam timing tool won't fit... worn cam pads, stretched chains, bad actuator, or other.

Your ign timing wasn't that far off, not sure if it's a reason to tear into it anyways???

Are there any other symptoms other than slightly down on power?

And you're certain your MAF is good and your resolved any intake leaks?

Could be the sensor too, as mentioned previously, you might be able to use a multi meter to check the resistance between pins to see if it is within spec. Or just try a different one.

Edited October 7, 2011 by logray

[mattatk]

Logray, I commend you on your effort, persistence & care....it is wonderful to have an expert around to discuss these issues. Thanks & I owe you a beer if you're ever in Australia!!!

When I get home from work I'll test the sensor & give it a good test drive to see if it's improved. You are right, it is still running well, I'm just overmaintaining her & noticed a slight burble on the idle & loss of overall power.

Thanks again