Questions about timing on VP Commodore V6

[silentshout]

Good morning,

Recently fitted a VP V6commodore with an 808 ecu NVRAM and High Speed board, we are running $12p firmware (Shoutout to all on this board responsible for developing it, its a really good bit of kit)

We noticed the car/timing felt very lazy compared to the factory VP commo ecu, Im guessing the VR timing on the $12p might be a very baseline tune aimed to get someone going safely.
I found a 92 VP bin and XDF in this forum and compared to the values we pulled from the NVRAM, it seems wayyyy more spicy than I thought it would be.

VP vs 12p difference.jpg

My question is if this difference seems correct (didnt think the timing would be that different for 0.5 compression ratio difference between the VP and VR) and if anyone has advice or experience on Buick / VP timing? Not looking for tunes more so known limits (knock limit or MBT) and other things to watch out for.

Car is running wildcat headers, Redback 2.5" exhaust out the back. Currently has stock intake with one of those Cold air over the radiator thingies into the stock box from back in the day.
If the difference looks correct then ill just slap those values on for the most part (and interpolate to the upper limit on the VR table which has another 4 RPM rows) as the baseline for out needs.

At the moment we are tinkering with ecu tuning on the NA setup but the goal is to rebuild a mang and then turbo it eventually.

many thanks in advance,

[vlad01]

I never really paid much attention to the factory maps on the V6 despite tuning them for nearly 2 decades in some form.

I never found any gains at all deviating from the factory maps in the engine's stock config, in fact they always lost power even increasing the timing a little (or reducing for the fact), as once proven to a lot of us early members where we did a in person meet up and dyno learning day. We put a stock VN up on the dyno, did a bunch of base line runs. Pulled around the stock 90-95 rw KW, we loaded up a HSV cal which comes from the SV3800 and various other HSV v6 of the time. In the map they smoothed out the map and increased it by a few deg. It actually lost power, it was down across the board down to low 80s KW.

So that confirmed that the factory cal is indeed at least with spark is spot on and MBT. This also pointed to HSV not knowing what they were doing with the V6 of that generation. But I will give them credit for figuring out the absolute ideal lean cruise AFR for these engines of 16.0. I did a lot of analysis on the lean cruise a few years ago, it needs 6% more timing in the corresponding spark area to gain an advantage and all lean AFRs from 15 to to 18 was tested in 0.1 increments with multiple runs over each on exactly the same section of road, speed, data logged, weather picked for no breeze, steady temps etc... you know, to rule out other factors, be scientific etc.

But I digress,

I have only bothered with tuning once I have long tube headers which in all the years of specializing only in these v6 engines, have found Pacemaker to be the only decent option that works, all other brands have primary tubes that are way too short to even do anything over stock. Pacemaker allows some real good gains to be seen, and I found from VR to VP/VNII to be around 5% difference in timing. The map remains exactly the same shape, but the lower comp demands that approximately 5% more timing.

Both engines are on the verge of being knock limited at their MBT. My tunes are based on actual dyno MBT tuning, across the entire map. Many hours were spent dialing in the map for these engines running the long tube Pacemaker.

Of course the header is not the only thing that is needed to gain power, but it is the thing that I found determines the spark map in its entirety. Added air flow from say a better air box only changes the accessible VE areas on the map, for eg. the stock airbox restricts you to the high 80 to the low 90 Kpa area of the map at sea level, the much better Mace airbox allows you to get that 100Kpa or at least very close to (~99Kpa) that full atmospheric pressure.

With the knock limiting. It does not occur in 99.9 or whatever % on regular fuel, on 98 it basically never happens except at a almost unnoticeable level in that 0.1% of the time. On 91, it's still that 0.1% of the time but much more obvious when it happens. I found it only occurs in two places in Melbourne out of all the places I have driven in Vic, I don't live in Melbourne, rather I was a field engineer tech that traveled all over and I used to do 300km on average in and around Melbourne and back each day, so drives were repetitive daily for almost a decade, great for data!

I found the condition to be low gear usually 1st or 2nd, light load around 10% throttle, medium rpm around 3k, slow speed 20-30km but constant and no start stop, in air restricted traffic. The two areas are the tunnel on citylink outbound on the hill up, and the other is the long circular bend on the calder/tulla bound getting up onto the bolte. Both spots have this constant medium rpm, slow hot airflow, long duration airflow. I found that being close to sea level is also a prerequisite for this rare but repeatable knock event to occur. The knock is actually a burst knock event, it happens when you are finally allowed a new lane to pass all of the shit slow moving trucks and other moron car drivers and blast past them, that sudden acceleration after that long drawn out rpm hold in no front air draft condition allows the air temps to get quite high and the sea level allows charge pressures to tip it over into this knock limited scenario.


In one of my cars where I did all the upgrades but was still waiting for a Mace box, I ran a straight pot on the end of the duct, and it also started to get this from traffic light launches (only in Melbourne). So confirming that is it a combo of air temp and high atmospheric pressure.

Due to the knock control system on these Delcos being totally useless, I found that it doesn't pick up on real knock, and simultaneously picks up on random nothing, can't even say false knock as there is literally no sound source from anything to be triggering these random events. Some cars are worse than others in this regard. One car I had years ago just kept losing power over time, I ruled everything out, when I finally realised it was the knock retard was activating at random it seemed and pulling the full 12 deg as set per factory. This was many years ago and was the major factor to moving on to a consistent performing tune for all my V6s going forward. That example car, I started off with getting around 7.5-8 sec to 100km/h, then it slowed down to around 9.5-10 sec after a few years of driving. I though the engine was just tired, but after finding that out, I disabled the knock retard and the car went back the 7.5 sec and was consistently quick right to the end.


I found that with these engines knock limiting to be a pre condition rather than simply a matter of too much timing, the timing needed to be pulled from the table to prevent this rare but repeatable knock limit is too much that it lands it back into the stock map range and thus losing the gains. And since the knock system is so hit and miss, yeah that won't work either.

I did a thread not long ago looking into using the flex table to use the air manifold air temp x tps or something to trigger an on/off output that feeds backing the map A/B switch input where it will use a lower timing B map until the air temp and TPS fall below a certain value again, returning back to the default map A. A hacked way of adding a air temp vs spark table, since 12P or any of the 808 family Delco lack in their code.


There is a charge temp spark subtraction table, but I found the calculated (derived) charge temp value does not correspond too well to these conditions, that it won't work for preventing these knock events, the charge temp around this condition is mostly the coolant contribution, not the air temp. You can't mess with that to make it more of the air temp without messing all other fueling functions up.

12P really needs a spark vs MAT subtraction table, so my work around is the flex table feeding back into the ECU. I have not tried this yet, but the consensus is that it should be possible and work.


For most people, this won't matter, also 98 fuel pretty much mitigates this fully. But I like technical engineering problems and I would also like to be able to use 91 fuel as it isn't knock limited on these engines fully tuned except for that one rare and unlikely scenario. But for the purpose of this thread, it is a good illustration of how the knock limiting on these engines work, it's more complex and dynamic than most people think, it's not a simple too much/too little timing, or safety margin approach like most tuning advice given out.

I will also add that with the stock cast logs or short tube headers as you would have, the knock limiting gets much worse if you push the timing, but there is no need to as mentioned, the factory timing is already at MBT for the stock logs or any of the short tube headers out there. Just as a FYI, the Pacemakers are just barely long enough for these engines to be ideal, but they are in the length range where their gains are in the upper/plateau part of the gain curve, anything shorter basically falls of the cliff into the rage that stock manifolds reside in. So in other words, Pacemaker work good, but only just by length, anything shorter falls off to bugger all gains. And everything else on the market is way shorter, like by half, sometimes less, only slightly longer than stock. There is a graph of this length vs gain curve online that came from David Vizard, it applies to pretty much any engine and basically shows that primary length is much like a long enough is good enough type of deal, and the lengths before you reach that good enough are much longer than most people think they are.

I tuned a hot VR engine in a VN about 18 months ago, it had the cheap short headers in it, the cam wasn't great (stage 3 crow cam), but still didn't account for the poor performance, it was almost no more power than stock. I blame the headers letting the rest of the engine combo down.

The ideal primary length for these engine in their stock form (stock cam) is about 1100mm, the Pacemaker is around 650-700mm. Some very wild cams might demand down to 850mm, but that's as short as the maths says they get in the ideal world.

Bit of an info dump, but I figured this is right up my alley.

[Charlescrown]

I look at the factory maps and get confused with the wide variation in data but when you consider they interpolate it may well cover/correct what looks like sloppy tuning. Interesting to read vlads comments about tuning and getting little if no improvement. I do know that with an engine there are many factors affecting the timing and fuel needs and would think GM spent huge amounts of time and money getting it right. I guess we will never know why they did what they did. The old carbie did it by the air flow thru the venturi and timing determined by springs, weights and vacuum. Very wide/varied and sloppy but met emission regs.

[antus]

There are 2 tunes that come with 12p. Series 1 (APNX) and series 1.5+ (BLCD). I suspect you might have the wrong one for the engine loaded. You can just load the non-stacked version in tunerpro and hit upload to send it in, or use the delco ose flash tool to upload the cal. But you can also do your compare in tunerpro first to see if that is what is going on. You may have APNX in there instead of BLCD.

[delcowizzid]

Vn vp and vr all had different spark curves .copy over the vp table colomns you can and interpolate the columns that arnt there

[immortality]

I look at the factory maps and get confused with the wide variation in data but when you consider they interpolate it may well cover/correct what looks like sloppy tuning. Interesting to read vlads comments about tuning and getting little if no improvement. I do know that with an engine there are many factors affecting the timing and fuel needs and would think GM spent huge amounts of time and money getting it right. I guess we will never know why they did what they did. The old carbie did it by the air flow thru the venturi and timing determined by springs, weights and vacuum. Very wide/varied and sloppy but met emission regs.

I had some some good gains playing with timing and fuel on my VN. Once you start playing with things like exhausts and intakes things change and so should the tune.

I don't think Holden ever tuned for peak power, they tune to achieve a set power target whist meeting EPA and NVG limits, I don't think they are ever optimised for peak performance.

People also forget about the idle to mid range, to many people focus only on peak power and WOT.

edit: I should probably qualify that by saying that here in NZ we had different tunes to aussie, we were still on leaded fuels when the VN's came out here.

[vlad01]

Oh yeah and I forgot to mention, the bellmouth insert in the manifold only causes issues. Gareth and I did a back to back dyno with and without and also ruled out other aspects of the tune that might skew the data, but consistently, the bellmouth killed power across the entire rpm range and nuked the torque in the mid to low the most, it also made the torque curve wavy with a few larger dips. Without was the best without question.

On this particular engine, it ran a crow stage 3 (very mild cam with poor LSA specs), it lost around 11Nm of torque on average by adding the bellmouth back in, some areas were a lot greater due to the dips mentioned whereas without it, the curve was the typical mound shape with not too much wiggles.


Also, APXX is the VN LN3 tune, that engine has a different cam and bit of a different design overall, cam has a wide LSA and slightly wider lobes, it responds differently to the later ones and I found doesn't gain as well to headers, much like the cams from crow which are almost all wide LSAs too, these engines like a tight LSA and the later VN to VR ran a slightly smaller but fair bit tighter LSA, so the headers makes them gain a bit better despite having a smaller cam on paper. Due to these differences, the VN series 1 naturally ended needing more timing and has so from the factory.

[antus]

Vn vp and vr all had different spark curves .copy over the vp table colomns you can and interpolate the columns that arnt there

they were but I remember VL400 had done comparison and testing and said they all ran well with the later map except the series 1 VN which is why there is also APNX.

[silentshout]

I never really paid much attention to the factory maps on the V6 despite tuning them for nearly 2 decades in some form.
...
Bit of an info dump, but I figured this is right up my alley.

Wow that is a treasure trove of valuable information and quite the read thank you so much, makes alot of sense. for the past year ive been tuning my Hyundai Tiburon and in that case Hyundai were exceptionally lazy with the timing but I think they were making that engine (V6 also found in some of their SUVs) to work on the most 3rd world dinosaur juice available. interesting there were tuning differences between regions, EU being the most restrictive for obvious reasons.

There are 2 tunes that come with 12p. Series 1 (APNX) and series 1.5+ (BLCD). I suspect you might have the wrong one for the engine loaded. You can just load the non-stacked version in tunerpro and hit upload to send it in, or use the delco ose flash tool to upload the cal. But you can also do your compare in tunerpro first to see if that is what is going on. You may have APNX in there instead of BLCD.

First thing I checked, before we plugged it all in etc, I noted the differences between the two, the wild difference in timing between the VN APNX timing and the VR BLCD stuff, one of the first things I did after plugging it all in was pulling the BIN file off the ecu to double check, its the BLCD one.
When we were test driving we felt power down nearly everywhere, also noted it idled far far better and there were some decel burbles that arent present on the VP ecu.

So I went hunting for the VP timing to verify and across the board its more spicy than the VR timing, my guess was compression ratio differences but could also be everything turned down to make a safe baseline tune.

I look at the factory maps and get confused with the wide variation in data but when you consider they interpolate it may well cover/correct what looks like sloppy tuning. Interesting to read vlads comments about tuning and getting little if no improvement. I do know that with an engine there are many factors affecting the timing and fuel needs and would think GM spent huge amounts of time and money getting it right. I guess we will never know why they did what they did. The old carbie did it by the air flow thru the venturi and timing determined by springs, weights and vacuum. Very wide/varied and sloppy but met emission regs.

I think part of this as well is knowledge around tuning has advanced since those times. Back in the day everyone hammered on about AFR this and AFR that to make the most power, modern tuning understands that timing, MBT etc are all arguably more valuable to how the engine operates and reaching the most volumetric efficiency for the combustion cycle.

Vn vp and vr all had different spark curves .copy over the vp table colomns you can and interpolate the columns that arnt there

Neat! Well you are a "delco wizard" so inclined to believe you This is what i was going to do for the large part save the low end stuff (20mg column) as the idle and decel are really nice on the BLCD out of the box settings. just dont know if i should be replicating the VP factory fueling as well or if the VP timing on the BLCD $12p everything else is safe enough.

Ill probably just start with 50% of the difference and bump it up from there to be safe.

Oh yeah and I forgot to mention, the bellmouth insert in the manifold only causes issues.

Good point! Forgot that might still be in there, should take it out when we can. Im thinking the bellmouth was to increase velocity low down increasing lower end torque but if your tests shows it does nothing then nfi what Holden were thinking

Vn vp and vr all had different spark curves .copy over the vp table colomns you can and interpolate the columns that arnt there

they were but I remember VL400 had done comparison and testing and said they all ran well with the later map except the series 1 VN which is why there is also APNX.

It ran well, as in smooth and nice, but it was severely down on power everywhere compared to the factory VP ecu, i thought it was just me so plugged the VP ecu in and mate drove it home and felt the same thing, VP ecu = more peppy / power but not as smooth idle and decel wasn't as nice


Cheers all, will see what we can do with bellmouth out and ill put 50% of the difference in timing on see how that goes and bump it up from there.

[vlad01]

I can't think of a legit reason on what the bellmouth was for, not only did it hurt power, but stuffed the air distribution to the cylinders too. Middle cylinders (3 and 4) run ok, front two (1 and 2) run much leaner, rear most ran too rich. Once the bellmouth is removed the AFRs balance back out pretty good. The difference is more than big enough to see it in the plugs, front white, middle tan, rear dark brown/black.

They all return to tan assuming the tune it right, all even anyway.

Once you remove the bellmouth, the VE around the light throttle areas and a few spots will increase slightly, some small spikes move or flatten out. It's def all benefit and no down sides that I could find.

The urban myth was to restrict the low down power and response to make the car feel easy to drive, more suitable for all customers including older drivers that liked smooth cars. This is mostly the reason the VN series 1 had a rep of being more powerful and responsive, peak power it was actually a little less than the series 2, but down low and off idle it was much better, most likely from this lack of bellmouth.

I am starting to think that maybe this myth has something to it.