VEMS install Porsche 911 Carrera 3,2 170kW/231PS motorcode 930-20, help needed
08-17-2011, 03:46 PM
#1
Track Day
Thread Starter
Join Date: Dec 2003
Posts: 15
Likes: 0
Received 0 Likes
on
0 Posts
VEMS install Porsche 911 Carrera 3,2 170kW/231PS motorcode 930-20, help needed
Hi,
i´m planning on installing a VEMS unit on a stock 89 Porsche 911 Carrera 3,2.
From what I read so far I´m quite confident that it should work, although there are no applictions so far ?
However some questions are left.
Let me start with the facts found so far and my plans for application:
- Motor 3,2litre normally aspirated
- trigger system is 2 VR sensors on flywheel / starter ring gear
- rev sensor is sensing 129teeth per crank rev
- TDC sensor is sensing 1 pin on flywheel (don´t know yet how many degrees before TDC)
- no cam or distributor sensor
- stock NTCII at cylinder head #3
- AFM (flap type) will be removed, will use TPS instead (exchanging stock idle and WOT switch)
- idle control valve is the "silver cigar" BOSCH thing, not sure if stepper motor or PWM ?
- IGN distribution will remain untouched in first step, stock coil, stock distributor
- stock fuel injector BOSCH 0-280-150-158 low-Z (2-3 Ohms)
- stock fuel pressure is 2,5bar
- wideband O2-sensor will be used intead of stock narrowband
- no MAP, no airtemp sensor planned
- Firing order for the Carrera is 1-6-2-4-3-5
- wiring should remain mainly stock with stock-DME-connector and only additional wires for wideband, TPS and so on
- planning to use alpha-N only
Now to the open issues:
- first step is to use as much stock as possible, get it running and then in the next step think about wasted-spark, sequential injection,...
- primary trigger: 129 is quite hard to divide, only possible value is 43 ! -> 6 triggers, 120° Will this work ? (AUDI div by n)
- any drawbacks due to the 129 teeth variant, like rev limit or so ?
- secondary trigger is the TDC-mark on flywheel
- what type is the idle control valve ?
- what to do for the low-Z-injectors ? Is there a bolt-in compatible high-Z-variant ? The stock setup is 3 injectors parallel for each bank.
The ones from the 964 are high-z but have lower cc/min @higher pressure
I´m looking forward to get into discussion with you guys.
Is there someone running a VEMS or MS on a 3,2litre ?
Best air-cooled grettings from Germany,
Marcus
i´m planning on installing a VEMS unit on a stock 89 Porsche 911 Carrera 3,2.
From what I read so far I´m quite confident that it should work, although there are no applictions so far ?
However some questions are left.
Let me start with the facts found so far and my plans for application:
- Motor 3,2litre normally aspirated
- trigger system is 2 VR sensors on flywheel / starter ring gear
- rev sensor is sensing 129teeth per crank rev
- TDC sensor is sensing 1 pin on flywheel (don´t know yet how many degrees before TDC)
- no cam or distributor sensor
- stock NTCII at cylinder head #3
- AFM (flap type) will be removed, will use TPS instead (exchanging stock idle and WOT switch)
- idle control valve is the "silver cigar" BOSCH thing, not sure if stepper motor or PWM ?
- IGN distribution will remain untouched in first step, stock coil, stock distributor
- stock fuel injector BOSCH 0-280-150-158 low-Z (2-3 Ohms)
- stock fuel pressure is 2,5bar
- wideband O2-sensor will be used intead of stock narrowband
- no MAP, no airtemp sensor planned
- Firing order for the Carrera is 1-6-2-4-3-5
- wiring should remain mainly stock with stock-DME-connector and only additional wires for wideband, TPS and so on
- planning to use alpha-N only
Now to the open issues:
- first step is to use as much stock as possible, get it running and then in the next step think about wasted-spark, sequential injection,...
- primary trigger: 129 is quite hard to divide, only possible value is 43 ! -> 6 triggers, 120° Will this work ? (AUDI div by n)
- any drawbacks due to the 129 teeth variant, like rev limit or so ?
- secondary trigger is the TDC-mark on flywheel
- what type is the idle control valve ?
- what to do for the low-Z-injectors ? Is there a bolt-in compatible high-Z-variant ? The stock setup is 3 injectors parallel for each bank.
The ones from the 964 are high-z but have lower cc/min @higher pressure
I´m looking forward to get into discussion with you guys.
Is there someone running a VEMS or MS on a 3,2litre ?
Best air-cooled grettings from Germany,
Marcus
08-20-2011, 03:08 PM
#3
Addict
Perhaps nobody here or on pelican understands why you want to do this. I don't get it. Are you planning future upgrades that will eliminate your current intake and fuel injection? Are you going to add a turbo down the road? I can think of a dozen other ways to add horsepower if that is all you are after.
Is this what you are talking about:
http://www.vems.us/node?page=1
http://www.vemssupport.com/forum/ind....html#msg18274
Is this what you are talking about:
http://www.vems.us/node?page=1
http://www.vemssupport.com/forum/ind....html#msg18274
08-20-2011, 05:58 PM
#4
Addict
08-21-2011, 03:09 PM
#5
Track Day
Thread Starter
Join Date: Dec 2003
Posts: 15
Likes: 0
Received 0 Likes
on
0 Posts
Actually I want to give my car a state of the brain:
As it only has one ECU, the DME, this one should be not outdated.
The main reason to build a platform for further tuning steps and to get rid of the following problems:
Lambda-control seems to be to slow causing
mid-rev shudder
slow throttle response
....
From what I see from various projects with other engine management systems it should solve all this.
There is a company called LENZ in Germany and I also see some guys using KMS from the Netherlands with great success.
But as VEMS is IMO the best-value DIY-system thats the way I plan to do with as much stock components as possible.
Maybe thats some project other can copy for as less money as possible !
Best regards,
Marcus
As it only has one ECU, the DME, this one should be not outdated.
The main reason to build a platform for further tuning steps and to get rid of the following problems:
Lambda-control seems to be to slow causing
mid-rev shudder
slow throttle response
....
From what I see from various projects with other engine management systems it should solve all this.
There is a company called LENZ in Germany and I also see some guys using KMS from the Netherlands with great success.
But as VEMS is IMO the best-value DIY-system thats the way I plan to do with as much stock components as possible.
Maybe thats some project other can copy for as less money as possible !
Best regards,
Marcus
08-21-2011, 03:45 PM
#6
Addict
Actually I want to give my car a state of the brain:
As it only has one ECU, the DME, this one should be not outdated.
The main reason to build a platform for further tuning steps and to get rid of the following problems:
Lambda-control seems to be to slow causing
mid-rev shudder
slow throttle response
....
From what I see from various projects with other engine management systems it should solve all this.
There is a company called LENZ in Germany and I also see some guys using KMS from the Netherlands with great success.
But as VEMS is IMO the best-value DIY-system thats the way I plan to do with as much stock components as possible.
Maybe thats some project other can copy for as less money as possible !
Best regards,
Marcus
As it only has one ECU, the DME, this one should be not outdated.
The main reason to build a platform for further tuning steps and to get rid of the following problems:
Lambda-control seems to be to slow causing
mid-rev shudder
slow throttle response
....
From what I see from various projects with other engine management systems it should solve all this.
There is a company called LENZ in Germany and I also see some guys using KMS from the Netherlands with great success.
But as VEMS is IMO the best-value DIY-system thats the way I plan to do with as much stock components as possible.
Maybe thats some project other can copy for as less money as possible !
Best regards,
Marcus
Let me go through the list of things I did and tell me if you have done the same.
1. Ignition wires, spark plugs, distributor cap and rotor. This helped a lot, but it still wasn't there.
2. I pulled the intake and put new fuel lines, head temp sensor, and intake gaskets. I think I had a small air leak. I also adjusted my valves about this time.
3. I did screw around with the cat-bypass and open air box cover, and put on a new O2 sensor, but gave away the bypass and box cover when I needed to get the smog testing done.
4. Then I did something that caused all kinds of change. I put in an aftermarket chip. This was like adding jet fuel to a fire. If you PM me I'll give you the details on the evolution of my chip. I now have a rather sweet setup custom mapped for the octane I can get.
5. I have replaced the WOT, TPS, and various tidbits, but nothing cured the 911 mid-range problems like the chip.
My suggestion here on the forum board is to go with a Steve Wong chip tuned to your exhaust and octane that can be picked up at www.pelicanparts.com . Steve will do map changes free of charge if needs be. It will dramatically change the throttle response of your engine. I don't know of anyone that isn't happy with there Steve Wong chip. I had one in my car for an evening and fell in love with it. I did have to give it back, and am very happy with the setup I have now.
Honestly, I can only think of one rennlister that has an aftermarket engine management system and aftermarket MAF. Everyone else here does it the old fashioned way. . . with chip.
08-22-2011, 02:53 PM
#7
Track Day
Thread Starter
Join Date: Dec 2003
Posts: 15
Likes: 0
Received 0 Likes
on
0 Posts
Actually I did most of it and I´m using a chip from Steve already.
The problem I described is what nearly every 3,2l driver tells me and my personal opinion is that the old 911-DME is near its runtime limits to calc the lambda control.
And, yes of course its a personal science project for me and will probably be a nice time to figure all this out.
The problem I described is what nearly every 3,2l driver tells me and my personal opinion is that the old 911-DME is near its runtime limits to calc the lambda control.
And, yes of course its a personal science project for me and will probably be a nice time to figure all this out.
Trending Topics
09-06-2011, 10:01 AM
#8
Marcus,
Most aftermarket ECUs don't like the 3.2L trigger with 129 teeth. My advice is you install a 60-1 trigger wheel at the crank.
The Motronic DME actually takes each of the 129 teeth and divides them into 4 so internally it sees 129*4=516 - it sees 516 pulses per rev. then if you take 360/516=0.6977 - so each pulse is worth 0.6977 degrees of ignition.
Most modern ECUs won't adapt easily to the stock pulses.
With all that said, a properly running 3.2L should not exhibit any performance issues. If you are seeing hesitation, throttle response issues you have something wrong. You really need to get a WBO2 system monitor on that engine to see what's happening to AirFuelRatios under these conditions. Often a common issue is the AirFlowMeter, these meters are getting old and the wiper arm track within them often gets worn and you get a dead spot in the signal. You can test this with the key in 'RUN' engine off then measure the output voltage from the AFM (0-5vdc) then using a pencil eraser side push the barn door within the AFM open while observing the output voltage. The voltage must sweep cleanly from 0-5vdc.
I have also seen faulty O2 sensors really screw up the mixture, I also suggest you disconnect the O2 sensor and see if the hesitation goes away.
Best of luck.
Most aftermarket ECUs don't like the 3.2L trigger with 129 teeth. My advice is you install a 60-1 trigger wheel at the crank.
The Motronic DME actually takes each of the 129 teeth and divides them into 4 so internally it sees 129*4=516 - it sees 516 pulses per rev. then if you take 360/516=0.6977 - so each pulse is worth 0.6977 degrees of ignition.
Most modern ECUs won't adapt easily to the stock pulses.
With all that said, a properly running 3.2L should not exhibit any performance issues. If you are seeing hesitation, throttle response issues you have something wrong. You really need to get a WBO2 system monitor on that engine to see what's happening to AirFuelRatios under these conditions. Often a common issue is the AirFlowMeter, these meters are getting old and the wiper arm track within them often gets worn and you get a dead spot in the signal. You can test this with the key in 'RUN' engine off then measure the output voltage from the AFM (0-5vdc) then using a pencil eraser side push the barn door within the AFM open while observing the output voltage. The voltage must sweep cleanly from 0-5vdc.
I have also seen faulty O2 sensors really screw up the mixture, I also suggest you disconnect the O2 sensor and see if the hesitation goes away.
Best of luck.
09-06-2011, 04:54 PM
#10
Addict
09-06-2011, 05:24 PM
#11
Keep in mind that it's not generally a good idea to switch to another ECU strategy if the current one is not working. You really want to be sure you don't have something else wrong because if you do you may just waste a bunch of money on very expensive ECU and still have same issues.
The stock DME is perfectly capable of correctly running these engines without any issues. A modern ECU may have some benefits but don't expect miracles. Bottom line, if you can't get the engine to ingest more air you won't get more torque or HP it's really that simple. And a ECU alone won't help you ingest more air, it will help you manage fueling that air and firing ignition and these 2 things can't be improved much beyond the stock setup.
Now if you tell me you made a bunch of mechanical changes like larger cylinders, more aggressive cams, better flowing intake and exhaust then these COULD improve intake air flow and now you may be at the limits of the stock AFM. But even here we have ways to handle these things with the stock DME.
The stock DME is perfectly capable of correctly running these engines without any issues. A modern ECU may have some benefits but don't expect miracles. Bottom line, if you can't get the engine to ingest more air you won't get more torque or HP it's really that simple. And a ECU alone won't help you ingest more air, it will help you manage fueling that air and firing ignition and these 2 things can't be improved much beyond the stock setup.
Now if you tell me you made a bunch of mechanical changes like larger cylinders, more aggressive cams, better flowing intake and exhaust then these COULD improve intake air flow and now you may be at the limits of the stock AFM. But even here we have ways to handle these things with the stock DME.
10-03-2011, 06:22 AM
#12
Three Wheelin'
I just stumbled to this thread.
There is nothing wrong with installing VEMS on the 3.2 Carrera. All of the sensors will work fine, but you will have to install air temperature sensor into the intake because you will loose your current one with the removal of the AFM (it is inside it).
129 teeth on the flywheel and 1 TDC/reference pin will work OK with VEMS, no problems. You are right about the divider, next trigger tooth in VEMS configuration is 43.
Regarding poor running, you will see a malfunctioning sensor immediately with VEMS and it is easier to diagnose issues. You have to replace the on/off type TPS with the one that has variable resistance and any BOSCH one from for example Opel Omega A (2.0) will work.
Stock idle valve is PWM type and works OK.
Having properly tuned standalone instead of ancient batch fire stock Motronic will make your car much better in terms of driveability and it also makse more power.
I installed VEMS on stock 964 Carrera 2 and my friend visited a dyno just before the install with the stock ecu and after. Difference was 156 rwkW vs 174 rwkW. NO changes besides removing AFM was made and his car worked perfectly well with Motronic otherwise. Properly tuned VEMS made it just better
There is nothing wrong with installing VEMS on the 3.2 Carrera. All of the sensors will work fine, but you will have to install air temperature sensor into the intake because you will loose your current one with the removal of the AFM (it is inside it).
129 teeth on the flywheel and 1 TDC/reference pin will work OK with VEMS, no problems. You are right about the divider, next trigger tooth in VEMS configuration is 43.
Regarding poor running, you will see a malfunctioning sensor immediately with VEMS and it is easier to diagnose issues. You have to replace the on/off type TPS with the one that has variable resistance and any BOSCH one from for example Opel Omega A (2.0) will work.
Stock idle valve is PWM type and works OK.
Having properly tuned standalone instead of ancient batch fire stock Motronic will make your car much better in terms of driveability and it also makse more power.
I installed VEMS on stock 964 Carrera 2 and my friend visited a dyno just before the install with the stock ecu and after. Difference was 156 rwkW vs 174 rwkW. NO changes besides removing AFM was made and his car worked perfectly well with Motronic otherwise. Properly tuned VEMS made it just better
05-15-2014, 06:12 AM
#13
Track Day
Thread Starter
Join Date: Dec 2003
Posts: 15
Likes: 0
Received 0 Likes
on
0 Posts
I have my 911 running with VEMS.
Now it is time to do the fine tuning.
Driveability and throttle response is almost perfect.
I got my issues with warmup and idle.
Any chance we can compare the configs ?
Can anybody decode all the stock motronic tables ? I allready have all standard tables, but I don´t have all correction tables.
Greetings Marcus
Now it is time to do the fine tuning.
Driveability and throttle response is almost perfect.
I got my issues with warmup and idle.
Any chance we can compare the configs ?
Can anybody decode all the stock motronic tables ? I allready have all standard tables, but I don´t have all correction tables.
Greetings Marcus
05-15-2014, 10:41 AM
#14
I know all the cold start tables in the stock DME but before you mess with cold start you need to get warm idle correct. So here's those details.
At idle run your ignition like this, let's assume idle speed of 900RPMs
RPM - ign
800RPM - 20deg advance
900RPM - 7 deg advance
1000RPM - 0 deg advance
Hopefully you have a table just for idle ignition if not then you'll need to massage the PT Ignition table to accomplish what's shown above.
The idea is to set timing at 7deg advance at the target idle of 900RPMs. Then if the idle speed dips you start feathering in ign so that if it drops to 800RPMs you'd be at 20deg advance, this helps use ignition to stabilize idle speed. Then if idle goes up above the target you remove ign so that at 1000RPMs you'll be at 0deg. If this is done correctly you can achieve decent idle even without Idle Control Valve.
Then for fueling at idle simply target 14.0 to 14.2AFR these engines like to run a tad richer than the 14.7AFR.
Get your warm idle working first then we'll cover cold start.
Also, what injectors are you running? The stock ones or something bigger?
And what's the injector strategy?
Are you running Batch (all injectors fire at same time)?
And do they fire once per crank rev (2 times per cycle) or every 2 crank revs (once per cycle)?
I ask these questions because at once per crank rev the inj pulse times are very small (down near 1ms) and some injectors can be unstable at these low pulse widths. The stock injectors can maintain stability down to 0.8ms or so. Do you know what the injector pulse width is at idle?
Also, EFI systems have a injector dead time table and it must be set accordingly for each injector type, have you done this? If you are using stock injectors where did you get the dead times from? You MUST get your injector model correct first, otherwise you'll be chasing your tail later trying to tune around improperly modeled dead times.
Finally, I'm recommending you buy these 2 books written by Greg Banish, they are the best books I've seen that clearly explain the tuning concepts and process:
http://www.amazon.com/Greg-Banish/e/...736&sr=8-2-ent
At idle run your ignition like this, let's assume idle speed of 900RPMs
RPM - ign
800RPM - 20deg advance
900RPM - 7 deg advance
1000RPM - 0 deg advance
Hopefully you have a table just for idle ignition if not then you'll need to massage the PT Ignition table to accomplish what's shown above.
The idea is to set timing at 7deg advance at the target idle of 900RPMs. Then if the idle speed dips you start feathering in ign so that if it drops to 800RPMs you'd be at 20deg advance, this helps use ignition to stabilize idle speed. Then if idle goes up above the target you remove ign so that at 1000RPMs you'll be at 0deg. If this is done correctly you can achieve decent idle even without Idle Control Valve.
Then for fueling at idle simply target 14.0 to 14.2AFR these engines like to run a tad richer than the 14.7AFR.
Get your warm idle working first then we'll cover cold start.
Also, what injectors are you running? The stock ones or something bigger?
And what's the injector strategy?
Are you running Batch (all injectors fire at same time)?
And do they fire once per crank rev (2 times per cycle) or every 2 crank revs (once per cycle)?
I ask these questions because at once per crank rev the inj pulse times are very small (down near 1ms) and some injectors can be unstable at these low pulse widths. The stock injectors can maintain stability down to 0.8ms or so. Do you know what the injector pulse width is at idle?
Also, EFI systems have a injector dead time table and it must be set accordingly for each injector type, have you done this? If you are using stock injectors where did you get the dead times from? You MUST get your injector model correct first, otherwise you'll be chasing your tail later trying to tune around improperly modeled dead times.
Finally, I'm recommending you buy these 2 books written by Greg Banish, they are the best books I've seen that clearly explain the tuning concepts and process:
http://www.amazon.com/Greg-Banish/e/...736&sr=8-2-ent
05-15-2014, 01:03 PM
#15
Track Day
Thread Starter
Join Date: Dec 2003
Posts: 15
Likes: 0
Received 0 Likes
on
0 Posts
Hi,
thanks for your response. Don´t get my wrong. I´m able to cold start and let the engine warm up without stalling it. Warm idle is also allready OK, but not perfect.
I allready did the trick of supporting the idle by IGN table around idle rpm. Additionally the VEMS has the ability to control idle with IGN offset.
Maybe i´m too much in the details allready but i try to make it as perfect as possible.
So i was wondering if the MAT compensations and so on are allready causing problems.
Currently i´m trying to fine tune the system under almost constant conditions. (Engine temp >100°C and MAT 20-25°C)
So my main interest was to have all the factory tables to derivate the settings for my VEMS config.
here are the details of the installation:
http://www.vems.hu/wiki/index.php?pa...ersPage%2FMacB
Short summary, stock setup, except TPS and MAT instead of AFM, stock injectors, batch fire like stock
pulse width in idle is approx 3,8ms.
Injector dead time table was a mixture of try and error and some internet sources.
There are no seperate WOT or idle maps. So i put in a idle row at low TPS and WOT row for high TPS
The values of the IGN are based on the stock chip of a 930-20ROW (as this is my engine)
What i don´t know if the TPS % are directly related to the values in the stock tables or if it is totally wrong to transfer this.
The VE table is % of base pulse width (7ms). The correction tables are % of nominal value at given loadpoint.
Here are the screenshots of the tuning screens in VEMSTune:
http://www.vems.hu/wiki/index.php?pa...%2FScreenshots
thanks for your response. Don´t get my wrong. I´m able to cold start and let the engine warm up without stalling it. Warm idle is also allready OK, but not perfect.
I allready did the trick of supporting the idle by IGN table around idle rpm. Additionally the VEMS has the ability to control idle with IGN offset.
Maybe i´m too much in the details allready but i try to make it as perfect as possible.
So i was wondering if the MAT compensations and so on are allready causing problems.
Currently i´m trying to fine tune the system under almost constant conditions. (Engine temp >100°C and MAT 20-25°C)
So my main interest was to have all the factory tables to derivate the settings for my VEMS config.
here are the details of the installation:
http://www.vems.hu/wiki/index.php?pa...ersPage%2FMacB
Short summary, stock setup, except TPS and MAT instead of AFM, stock injectors, batch fire like stock
pulse width in idle is approx 3,8ms.
Injector dead time table was a mixture of try and error and some internet sources.
There are no seperate WOT or idle maps. So i put in a idle row at low TPS and WOT row for high TPS
The values of the IGN are based on the stock chip of a 930-20ROW (as this is my engine)
What i don´t know if the TPS % are directly related to the values in the stock tables or if it is totally wrong to transfer this.
The VE table is % of base pulse width (7ms). The correction tables are % of nominal value at given loadpoint.
Here are the screenshots of the tuning screens in VEMSTune:
http://www.vems.hu/wiki/index.php?pa...%2FScreenshots