GeneralDisorder

iTrader: (1)

Installed my EBL yesterday and tuning has been really easy. RBob - you have created an amazing product. Thank you!

During warmup - once it's on closed loop - my BLM's are rich. Around 114 at idle. And they slowly climb back up to 128 at full operating temp. I've read a few threads that suggest this is pretty normal (although they reported lean during warmup), and could be the result of variations in the O2 sensor when it's still on the cold side? I don't see any tables for O2 bias.

Also I have a strange valley at 3200 RPM where the spark advance wants to slope down from about 27 degrees @ 75 to 11 degrees @ 100 or I get quite a bit of knock through that area of the RPM range. The "normal" LB9 map has no valley here. I do have headers, full exhaust, and 1.6 rockers. Any thoughts? It feels fine even with the SA retarded through there and I haven't heard any detonation (but I doubt I would at 3200 either).

GD

GeneralDisorder

iTrader: (1)

Think I might be on the lean side up there at high MAP/RPM. Ran into that with the MAF calibration for the LB9 also. I guess the 1.6 rockers and headers are needing quite a bit more fuel compared to stock. Enough to lean it to the point of detonation.

GD

RBob

iTrader: (1)

The SA dip at 3200 RPM is normal. That is where the long tube runners start to work. It is also common to see a peak in the VE at that point.

RBob.

GeneralDisorder

iTrader: (1)

Good to know. I'll continue tuning with that in mind. Thanks!

GD

GeneralDisorder

iTrader: (1)

I noticed the warmup BLM's seem to be related to the manifold temp readings. The lower the manifold temp on startup, the lower the BLM's once it goes into closed loop. As the manifold warms up, the BLM's return to normal. I'm not so much concerned about the BLM's going low at cold start as I am about what will happen when ambient temps drive the manifold temp considerably higher this coming summer - I'm only assuming that's what will happen though. I don't really want the BLM's climbing to the lean territory when my manifold temps climb while idling in the summer heat. Is there a bias for this or would moving the temp sensor out of the plenum and into the intake (like most other IAT sensors I'm familiar with) then relearning the VE for overall lower temp readings be better?

GD

ULTM8Z

I'd advocate for relocating the IAT and then recalibration of the MAT tables.

that effort had a very big payoff for me.

GeneralDisorder

iTrader: (1)

I could see that stabilizing the manifold temp at least from engine heat - so you'll only have a swing from say 20 degrees to around 120 degrees (for me)... seasonally and of course between moving and stationary idle. But that should be much more accurate as to the real air temp and not biased by the hot manifold and plenum from the engine heat directly beneath and in contact through the runners.....

What were the payoffs for you? I can easily move mine to an open element IAT sensor in my intake plumbing. I'll put it where the MAF was and use the existing (no longer used with EBL) MAF wires to route the signal.

GD

ULTM8Z

the payoff for me was a noticesble seat of the pants increase in torque as well as a consistency in how it runs no matter what the ambient temperature is and no matter what the IAT reads. it just runs the same all the time.

GeneralDisorder

iTrader: (1)

I think it makes the most sense. I'm going to move mine also for the new engine going in with the First intake.

GD

Dartht33bagger

Where did you move the IAT to? Did you create a custom intake to the TPI throttle body?

ULTM8Z

to start, my car is a 2nd gen Camaro, so itll be a little different. but yeah I relocated the battery to the trunk which allowed me to route a custom 4" plastic tube to that corner of the engine compartment. I cut a hole in that area and mounted a flange on the sheet metal which then allowed me to attach a conical air filter underneath directly in front of the wheel.

I wanted to get the intake completely out of the engine compartment. I also put the IAT sensor right in the snout of the filter so as to not get any sensor heat soaking.

GeneralDisorder

iTrader: (1)

I think I'm getting a lot of false knock. I've got the SA retarded by 10+ degrees and I'm still seeing it. I don't hear any detonation, and the SA map is getting so low it's affecting performance at WOT. I've checked the data logs and my O2 is 950+ mv every time the knock counts increase so it's clearly not lean. I've read quite a few threads claiming headers will cause false knock also. I have Dyno Don headers..... Not sure what to do other than just eliminate the knock retard.

GD

ULTM8Z

might be a crazy idea, but while the engine is running and while watching the datalogger, try tapping the header collector (on the side nearest the knock sensor) with various objects (trying to generate various frequencies) and if you get any kind of response by the ECM, such as knock counts or actual retard.

there's another member here who's going through something very similar on his car.

jeffas000073

may want to compare notes with him...

ULTM8Z

might be a crazy idea, but while the engine is running and while watching the datalogger, try tapping the header collector (on the side nearest the knock sensor) with various objects (trying to generate various frequencies) and if you get any kind of response by the ECM, such as knock counts or actual retard. that could tell you whether its possible to generate false knock.

a positive result would say yes, though a negative result wouldn't necessarily mean impossible....

there's another member here who's going through something very similar on his car.

jeffas000073

may want to compare notes with him...

the other thing is sometimes rocker arms can be pretty noisy as well... you could potentially put the stock ones back on and see if the problem goes away.

another (slightly riskier option) is disable the knock retard, and then ease into the throttle at 3200 rpm and see if you can actually hear detonation starting.

ULTM8Z

lol... are you runni g NGK plugs?
see last post in this thread....

https://www.thirdgen.org/forums/diy-prom/356351-tuning-ebl-92.html#post6272660

GeneralDisorder

iTrader: (1)

Sounds a lot like what the guy in that last post was experiencing but I'm not running NGK plugs. I'm running shorty Accel plugs for header clearance.

I wonder what is different about the 85 ESC module. I'll have to look into that.

I'm going to check the timing and also make sure the latency on the large cap distributor module matches the table I found on a thread here. But I'm still positive it's false knock.

GD

GeneralDisorder

iTrader: (1)

Well I checked the timing - both static with the timing connector disconnected and then with the advance hooked up and watching the EBL advance being requested. Looks dead on to me. Even at 5000 RPM the advance on the balancer was within a degree of what was requested so it seems the large cap latency is correct, and the static position is correct. At least as far as the markings on my aftermarket balancer and timing pointer are concerned. Looks like it's time to verify TDC with a piston stop and see if the balancer is even marked correctly or my pointer is off.

I disabled all the knock retard entirely and I did hear some detonation.

One thing I haven't done is back down the timing and disable the burst knock. But I don't think burst knock can pull as much timing as I'm seeing get pulled. At times it's averaging 5 or 6 degrees. I have seen 8 degrees.

Question for Rbob - does the burst knock retard show up in the Spark screen of the WUD even though it's not real knock?

GD

RBob

iTrader: (1)

Burst retard will show up as a reduction in SA. Note that burst knock retard is for tip-in knock. It reduces the SA by a set value for a set period of time. There is an option flag to enable it.

The older ESC modules are typically a 4-pole filter. This makes them more discriminating and report less false knock. But they are also more expensive so in '87 (or so) GM changed to using 2-pole filtering. Module BLH is a good one to use, hot stamp number is: 16022621

RBob.

GeneralDisorder

iTrader: (1)

After looking over my datalogs I do seem to have some burst knock happening. My knock counts increase when the TPS is showing a sharp increase. The whole time this is happening I'm showing in the 950 mv range for my O2 sensor so it doesn't appear that I'm lean.

I read elsewhere where you stated that with the improved fueling calculations, burst knock retard really isn't needed. Since I have 1.6 rockers and headers perhaps I'm getting a lot more air on rapid throttle opening than the stock LB9 equipment.

What changes to the VE, AE, or PE should I make that would help with burst knock besides enabling the burst knock feature? I'm concerned that my fueling isn't setup right if I need to use it.

GD

Orr89RocZ

iTrader: (20)

GeneralDisorder

iTrader: (1)

I know right? I've got mine at 15 degrees right now too (14.77 in the table, and no adders) and have the same problem. I just checked and I have the BLH module already (86 TPI with large cap distributor) so that's a dead end.

I have added as much as 10% more AE. Didn't help.

I'm fixing to put in a modified 400 HP 350 Vortec that's ready to go on a stand in my garage. But I wanted to get the business of learning the tuning specific to the EBL system down before the swap.

Having never tuned anything this old, I'm struggling with the seeming inability to clear up this knock sensor behavior and there is no way to desensitize the sensor because of the ESC module being a separate component.

Really this problem has been here the whole time - before I did the headers and before I installed the EBL. When I was tuning on $6E with stock headers and stock rockers I had this problem. The knock sensor is new - properly torqued. Engine condition is excellent - 195 compression on all 8. Nothing is loose - engine torque doesn't cause the headers to hit anything..... I just don't get it.

I guess since these engines were largely carbureted with no kind of knock sensor system.... does it really matter if there's detonation that I can't hear? I'm used to newer engines - some of which are very sensitive to detonation as it will pound the rod bearings right out of them. Of course they have much more sophisticated knock sensor control systems....

GD

RBob

iTrader: (1)

Then there is real knock.

RBob.

ULTM8Z

what are the AE tables on the EBL? 7730 ECM has quite a bit of AE tables to mess with.

it may actually be that 10% richer isnt enough? or that it's not coming in fast enough?

one thing I've learned with years of playing with AE while tuning my miniram is to not tune based on what you think it "should" be. tune for what the engine actually wants.

GeneralDisorder

iTrader: (1)

Yes there are a lot of AE tables in the EBL. The AE is pulse width based and so needs to be scaled (down) for the injector size per the EBL instructions - mine are about 15% larger than stock (Bosch III 24lb, scaled at 22.5lb) Which initially I did scale the AE down by 15%, but then dropped it back to stock for 19lb injectors, and then added another 10% on top of that - so I have been as much as 25% more AE than the 305 TPI bin file that was included with the EBL. And as I said I had this problem before with the $6E map even before I did the rockers and the headers. I have never had stock injectors as the car came to me with a dead fuel system (sat for 11 years).

The tables that I scaled are the ones recommended by RBob for injector size scaling - AE - MAP PW, and AE - TPS PW. I haven't messed with any of the other tables. There are two tables that affect the duration of the AE and magnitude of the AE. These are "AE - MAP Filter" and "AE - TPS Filter". They have a coefficient based on coolant temp.

GD

GeneralDisorder

iTrader: (1)

When I was running a stock SA table and no knock retard at all I did hear some pinging. Not very pronounced. I dialed back the timing to 15 degrees at WOT and don't hear anything yet I'm still getting knock counts on hard throttle opening. I tried turning on the burst knock feature and it doesn't seem to make any real difference. I even raised the MAP and TPS threshold for burst knock and didn't see any change.

GD

GeneralDisorder

iTrader: (1)

Tonight I'm going to verify the timing marks on my balancer with a piston stop from my degreeing kit.

GD

RBob

iTrader: (1)

As for burst knock, there is an option flag to enable it:

Option Word 1 - Bit 0 - BstNk

Also, any change in the engine 'noise' can have knock reported. Example, totally stock '92 3rd gen, when rolling into the throttle and the auto trans downshifts. That alone has knock being reported.

Tire spin or even a bit of tire chatter will do the same.

RBob.

GeneralDisorder

iTrader: (1)

Yes I turned on the option word bit for the Burst Knock. Seemed like maybe it helped to a degree.

So with my hard shifting 700R4 (fully built, 3k stall), I should expect to see some timing getting pulled and I shouldn't worry about it? From the logs it's usually a few degrees and then it seems to put the timing back in pretty quickly....

And yeah from a stop if I get some tire spin that also seems to get me a few counts. It's not hard to get some barking with the 3.70's.

What are the performance implications of these false knock counts? If I were to take it to the track would I just turn off the knock control and only run it on the street? Or would it likely not make any difference at the track either way?

FWIW I carefully varified my balancer and pointer with a piston stop and both are dead on for zero at TDC. So my base timing is definitely at 6 degrees.

I've got my WOT timing at 15 degrees right now. The performance is definitely hurting from it, I can hear no detonation of any kind, and yet at 4000 RPM / 95 MAP I'm seeing 3-4 degrees peak knock retard. I wasn't paying attention but that easily could have been on a downshift. I'll pay more attention to what the trans is doing from now on.

I wondered about the torque converter stall speed being around 3000 and it does seem a lot of my knock counts are clustered around the 2800 to 3200 range. Probably right around where the RPM falls on an upshift too.... it's a non locking converter if that matters.

GD

ULTM8Z

have you tried loosening up the sensor in the threaded hole in the block?

I've heard overly tight install could produce some hypersensitivity on the sensor.

does it knock only when fully warmed up and in closed loop?

GeneralDisorder

iTrader: (1)

I don't usually drive it prior to being in closed loop. I can try that and see.

I did torque the knock sensor with a digital torque wrench to factory spec but it wouldn't hurt to loosen it and check the torque again. I'll do that also.

GD

RBob

iTrader: (1)

Can also mount the knock sensor on a 45° street elbow.

RBob.

ULTM8Z

the additional threaded interface helps to attenuate some of the other noises?

GeneralDisorder

iTrader: (1)

So this morning I did not get any knock counts till it got into closed loop. Didn't hear any detonation either (this was with the stock LB9 SA table).

I tried decreasing the PE AFR because I saw that it was 1.6 lower when cold..... and increased the AE but it made no difference.

Any thoughts?

GD

ULTM8Z

Will the EBL actually report knock in open loop if it occurred?

I guess I"m trying to see what things your occurrence of knock is related to...and then which one of those things we can manipulate to make it disappear/reappear. i.e., closed or open loop? coolant temp? CL or OL and coolant temp? Richer mixture as a function of lower coolant temp in OL? etc... For example, if we can drive it with coolant temp, then maybe a 180F thermostat and lower fan thresholds could make it go away (assuming you're running a 195F). Or perhaps if it's a richer mixture that helps, perhaps you adjust your O2 R/L thresholds a little richer at higher MAP's.

RBob

iTrader: (1)

Yes, it works to desensitize the sensor. Can also use a short straight stub if there is room. It will usually hit a stock starter, which is why the 45° elbow is recommended.

Yes.

RBob.

GeneralDisorder

iTrader: (1)

True. Rbob says it will report knock in open loop so that kindof indicates it may be a tuning issue after all. I saw no reported knock till it entered closed loop.

I am currently running a 170 thermostat with fans set to 185 on and 175 off.

Should I try disabling closed loop operation and see if that alone prevents the knock counts?

GD

ULTM8Z

Could try that. If it gets rid of the knock, it would be helpful to know what your actual average AFR is during that condition. If you're running a little richer commanded AFR during OL, it may help to cover an early lean tip in.

170F... yeah that's already pretty cold... a little too cold IMO... but that's a subject for a different day! lol...

If knock remains during OL operation after full warm up, you could then try to fatten up the OL AFR at the higher coolant temp and/or richenthe AE vs coolant tables (i.e., AE delta TPS vs coolant, AE delta MAP vs coolant, or whatever the equivalent EBL tables are.... I'm still speaking in $8D-eese)....

Street Lethal

iTrader: (16)

That is normal. The air is more dense at lower temperatures, and the air/fuel ratio is preset to compensate for it. A slow rise in BLM keeps the idle transitioning smoothly...

What do your plugs look like? Your plugs will underline if you're detonating or not.

- Rob

ULTM8Z

I dont know if there's anything in the EBL that's equivalent to the AE delta TPS scale factor in the 7730 (not the AE delta TPS PW scale factor, but just the scale factor).

if there is you may want to try increasing it a little. I'm not sure what its actual function is, but it seems like it brings the AE in faster and was really the single biggest factor in getting my Miniram dialed in on AE.