PDA

View Full Version : Ls3 swap questions



tor1965
June 17th, 2010, 07:33 AM
Hi
I have a Caprice modified as a impala ss clone with ls1 and 4l60e transmission. I have used the car winter and summer for 3 year with this combination. I am living in Norway, so winter means real winter with down below -30 degrees Celsius and a lot of snow. So drivability is an important issue.
I have felt that it is something odd with the way the engine act on full throttle all time. I have played with different os and different settings during this time without getting the problem solved.
The stock os is 344, but I have mostly used 156 and 074. I have tried auto ve and weatherman Shawn tutorial. My lc1 has showed values of 0.3 lower afr then the narrowband all time. Even, after I bought a brand new lc1. The ve table looks strange after my tuning attempts with al to low values in the full throttle part.
The engine has some annoying cold knock, and now I feel that it have a small leak in a head gasket also.
Because I use the car every day, and that the engine obviously needs a complete tear down, I now have a brand new ls3 gm performance parts crate engine in my garage. It may be a little overkill for my use but hopefully very fun.
My plans are to run this engine with the 2000 f-body computer and wiring harness.

The parts I have bought for the ls3 swap are:

19244097 gm performance parts crate engine
Lingenfelter TRG-002 58x to 24x conversion module
Racetronix RX-ls2-KSRH ls1 to ls2 knock sensor relocation harness
CE-109077 Injector Connector Conversion Adapter EV6 to EV1
Caspers 109103 ls1 ls3 map sensor adapter
Nick Williams 92mm cable throttle body
Harmonic balancer bolts

I have a big Corvette style aftermarket maf, and
A maf sensor adapter harness

Then some questions.
Should I buy two new o2 sensors for the engine to start with? The old switch ok but is it possible that it can be something wrong with them though.

The ls3 map needs different settings then the ls1, and they say that this can’t be done with the old 512 kb pcm, but under engine diagnostics I have found C6301 and C6302, map scaler and map offset. Can I put in the values from an ls3 tune here to adapt the ls3 map to the ls1 os?

What os is best suited for this application? I dream of an Edelbrock supercharger in the futures, but who knows. I suppose that all arguments to keep the stock 344 os go out with the old engine?

My pcm is a 09354896. Should I buy a 441 instead?

ScarabEpic22
June 17th, 2010, 07:54 AM
Your PCM can simply be full flashed with the best OS out there (12212156), the physical PCM is the same.

Im assuming you can either use the LS3 MAP settings in the LS1 PCM or use the LS1 MAP. Sorry I cant help more, but definitely use the 12212156 OS. Id use it now if I were you, but that'd be a bunch of calibration copying.

joecar
June 17th, 2010, 10:31 AM
+1...

run 12212156...

copy the all tables over that can be copied, it only takes a couple of hours... the other tables may be arrived at by calculation or tuning with a wideband.

12212156 allows a very migration into COS3 (02020003) or COS5 (02020005) when you later install the SC (...you don't have to wait until then...).

joecar
June 17th, 2010, 10:32 AM
Hi Tor,

Is the cold knock related to piston slap...?

How many miles on the O2 sensors...? If they have 60K+ miles (100K km) or look fouled then replace them... when they get older they still switch but more slowly (they don't necessarily trigger any DTC's).

tor1965
June 17th, 2010, 11:38 PM
Hi Joecar

I am not sure if the cold nock is piston slap alone. I definitively have some slight piston slap noise at mild acceleration on about 1500 rpm+. The most annoying noise is a bit clunkier and occurs just after start up and goes away for some time after a punch on the throttle. When I drive for about 15 minute the noises is almost gone. At full throttle acceleration it feels like the valve is floating before I reach the shift point for second gear and the log file doesn’t show anything unusual.
Compared with some corvette engine my engine has a lot more “engine sound” on idle and part throttle, also when warm.

I think the o2 sensors have about 45k miles.
So old sensors switches slower but still retain 0.8-0.9 volt when the mixture is rich?

It seams like two new sensors would be a good investment.

Then to the os question, it seams like you both like the 12212156 os very well.
Shall I copy over all the tables I can from the ls3 camaro 2010 os. And use the rest of the tables from the 12212156 camaro 2002 os?

The 2000 and 2002 transmissions are the same, aren’t they?

ScarabEpic22
I have done all that copying and have used the 156 os earlier. I just swapped it out last winter because I had a 074 os with less aggressive shift points that was more suited for the winter.
I am going to use the ls3 map settings in the 156 os if I don’t get some other advice.

joecar
June 18th, 2010, 02:59 AM
I am not sure if the cold nock is piston slap alone. I definitively have some slight piston slap noise at mild acceleration on about 1500 rpm+. The most annoying noise is a bit clunkier and occurs just after start up and goes away for some time after a punch on the throttle. When I drive for about 15 minute the noises is almost gone. At full throttle acceleration it feels like the valve is floating before I reach the shift point for second gear and the log file doesn’t show anything unusual.
Compared with some corvette engine my engine has a lot more “engine sound” on idle and part throttle, also when warm. Piston slap usually stops when the engine has some load. Your other noise may be lifter related. Noises are quite difficult to diagnose.


I think the o2 sensors have about 45k miles.
So old sensors switches slower but still retain 0.8-0.9 volt when the mixture is rich?

It seams like two new sensors would be a good investment.Yes, there response time slows down... new: 5+ cycles/second; old: 1-2 cycles/second or slower.



Then to the os question, it seams like you both like the 12212156 os very well.
Shall I copy over all the tables I can from the ls3 camaro 2010 os. And use the rest of the tables from the 12212156 camaro 2002 os? Yes, copy all copyable tables from the LS3 file, and the rest from a 2002 Camaro file and/or from your own LS1 tune file. You will need the LS3 injector tables interpolated to fit your LS1 injector tables, see attached.

You have the returnless fuel rail, right...? Make sure fuel rail pressure stays constant at 58 psi at load.


The 2000 and 2002 transmissions are the same, aren’t they?
...Yes, they are the same.

tor1965
June 21st, 2010, 08:30 AM
Hi
Thanks for the zip file.

As a car mechanic I think that the sounds in my old engine are piston slap, and in addition it may have to big clearance in the crankshaft big end bearings.
It also has a very small leak in one of the head gaskets that pressurize the cooling system and cause that I lose a little bit of antifreeze. I think that this may be the cause that the wideband show strange values some time. When I get the ls3 in the car I am going to disassemble the old engine and measure every clearance and find the cause of this noise. Then I will decide what to do with it.

The thing that I like least about this ls3 project is that I have to populate a new ve table. When I have played around with my ls1 I have always had the opportunity to put the stock table back when I have felt that my tuning was “out on the edge” and then had a car that was safe to use.

Because you all tell me that the 156 is the best os, I have found the last 156 os that I converted from the stock 2000 os for my ls1. I had played with auto ve on this but the ve table look odd. I copied in the stock 344 ve table again, and flashed it back in to my pcm.

I was on a long trip up in the mountains this weekend with my family and some friends. I made a long log file under various driving condition. My friend drives a Mercedes and I just had to show him how slow his car is. The engine was going ok, it felt’s strong and it was not loosing power at high rpm as it has done some times earlier.

I have attached this ls1 log made by black box logging. It needs the calculated pid’s to be added. I also attached the tun file.
The maf is an aftermarket (Pro M) corvette maf that I have made a maf table to match.
The engine is stock except for dual exhaust after the cats, and the air ducting with the maf. 28 lbs injector from a 98 ls1 is also added.

When looking at this log. When the ve table values is too high does that mean that the engine is pumping less air then it should? Or is it the injectors that give more fuel then predicted?
On the ls1 I have fuel rail with return line from each rail to an Aeromotive fuel pressure regulator that is not connected to the manifold vacuum.
The pressure is measured on idle with cold engine several times. I have a Granelli in tank fuel pump and the stock impala nylon fuel line and fuel filter. Is this a source of error? When I get the Ls3 in I am going to use the return less fuel rail. I have not decided where to place the pressure regulator yet.
If somebody have the time to look at this log and tune, I would be very interested in the opinion on the maf settings. The wideband are showing at least 0.4 lowers afr then the narrowband. Is this something to take into consideration when tuning?

joecar
June 21st, 2010, 01:11 PM
Interesting log... :cheers:

Seems to me that MAF is reading a little low, and this may be the reason for knock being present at load.

Wideband AFR seems a little strange...

Yes, the VE table looks a little strange.

25° advance at peak torque may be a few degrees too much.

You could increase shift pressures a little and decrease shift times a little.

Look at B3809, that is classic F-car, you can set it to more sane values; for example, 900, 2100, 3300.

tor1965
June 22nd, 2010, 08:45 AM
I like your chart. It is far better then the one I have used, until now.
I just trying to understand how this pids works together, and wonder.
What are the difference between dyncylair_dma and cylair_dma?

joecar
June 22nd, 2010, 10:30 AM
My current understanding:

GM.DYNCYLAIR_DMA
Internal PCM calculated cylinder airmass [g/cyl] based on VE.

GM.CYLAIR_DMA
Internal PCM calculated cylinder airmass [g/cyl] based on MAF.


More info: goto my profile (http://forum.efilive.com/member.php?10-joecar) and scroll down.

tor1965
June 24th, 2010, 06:56 AM
The reason I play around with my tune before I changes my engine is that the engine don’t perform as well as it should. And I am afraid that this is not related to the engine it selves.
After I have studied lots of logs, and have thought on when the engine act strange I think that the problems are fuel related. Most of my earlier loggings and auto ve were done just after engine warm up. Then when logging a longer trip in summer temperatures the values changes. I have a feeling that this might be because the fuel warms up.
I have a home made return fuel system with the regulator placed on top of the engine. I have also tried some different solutions for the EVAP system and found pretty much vacuum in the tank sometime. The engine is in an impala ss clone. The in tank fuel pump is a Granatelli hp pump, but it is mounted on a 91 caprice in tank module. That means that it just sits in the middle of the fuel tank without any baffles or the more modern “turbo system”.
My fuel gauge is wrong as on most of these cars. When it shows half it is less then ¼. I have found that on Impala ss forum there is a lot of people that have the problem that the lt1 engines acts strange when the fuel level is down on1/4. When they dragrace this cars they newer do it with less then a ¾ tank of gas. I think that my feeling that the car loses power is because my fuel pump drags air on heavy accelerations. It is also possible that the fuel pressure changes do to the heating of the fuel pressure regulator or vacuum in the fuel tank.
This entire theory and tuning attempt and most of the problems could probably be solved if I borrowed the fuel pressure gauge that I have on work. I could have used it for weeks if I wanted to. On the ls3 I am going to place the fuel pressure regulator under the car or use the corvette fuel filter with the regulator inside.

tor1965
July 1st, 2010, 08:13 AM
Hi
I have another question.
I am working with my 12212156 tun file for the ls3 swap.
When I am trying to scale the ls3 map sensor in table C6302 (map sensor offset) the decimal numbers are rejected. Am I doing something wrong or does not the 156 os support decimal values in this table. If not, is there some workaround for this? The ls 3 map values are “map sensor offset” -0.31 and
Map sensor scaler 128.125000

ScarabEpic22
July 1st, 2010, 09:11 AM
Have you set the tune tool to display the decimals out to your desired resolution? There's a button in the toolbar that lets you increase (and decrease) the number of visible decimal places, very useful. A cell might show 0 but be anything from 0 - .99 in it.

tor1965
July 1st, 2010, 09:57 PM
Decimal button! So easy and yet so impossible to figure out :doh2:

Thank you, Erik.

ScarabEpic22
July 5th, 2010, 05:08 PM
No prob, simple little things that you wouldnt ever bother to check end up messing with your head! Been there, done that!

Glad you're up and running. :D

5.7ute
July 5th, 2010, 05:51 PM
tor1965. I dont think you want to use the map sensor adapter with an ls3 map sensor due to different pin outs or something like that.(I cant remember exactly why I was told not to do this)I just used the ls1 map sensor with the adapter in the ls3's location. It needs a bit of sealant & a small bracket fabricated to hold it in but this has worked fine for me.

tor1965
July 29th, 2010, 04:05 AM
The engine is in the car and it has been driven for about 1500 Km. The stock ls3 map sensor readings seam to be ok with the settings from the 2010 Camaro tun file.
I have done some work on the ve table, and the car runs fine. I still have to change the o2 sensors, so I need to fine tune the ve table later. The areas of the ve table that look a bit strange are the part under 1200 rpm. When logging it gives high readings during driving, but if I adjust the ve table up I got bad idle after startup.
I would not use any time on this before I have the new o2 sensors so this can wait.
My part throttle shift times have changed because of the 92 mm Nick Williams cable throttle body. The area of the throttle blade is much bigger and the TPS reading is obviously very low on normal driving. Is it possible to correct this in the tun? Without changing all settings where throttle position is used.