Big injectors.
06-20-2006, 09:26 AM
#1
Senior Member
Thread Starter
Join Date: Oct 2001
Location: Timrå, Sweden
Posts: 930
Likes: 0
Received 0 Likes
on
0 Posts
Car: 1984 Corvette
Engine: Turbo 350
Transmission: 4L80E with TCI T-Com
Big injectors.
I have modified a HSR to fit under the Corvette hood and I am also going to run a single turbo. I am using the old exhaust and running N/A before I swap to the turbo headers.
I have been using the Corvette for about a week now and I have no major problems. I have done a few hours of tuning and it is running quite good now. The tranny just broke in my winter beater so the Corvette is my daily driver right now.
I wired the injectors like the stock TPI. Right side on one channel and left side on one channel, batch fire. According the my WB lambda I need to run it a bit rich at about 2500-3000 rpm. WOT timing is constant in that range. If I keep WOT lambda constant it starts to miss in that area, and then it runs fine above. I think it is because of some resonace in the fuel pressure resulting in different fuel pressure for different injectors. I am using 120 lb/hr injectors and fireing all four in the same fuel rail at the same time might be too much. I am going to re-wire the injectors to use one injector from each corner as a group. If I add fuel in that area the problem goes away and I think that it is because the 'lean' cylinders get enough fuel. The 'rich' ones are probably way rich. I also tested with another nr of spark pulses per injector pulse, that changes both the injector frequency and pulse length. When I did that the 'problem' rpm range also changed. This is the reson I suspect some resonace in the fuel pressure.
I probably need to add a fuel pressure damper too, something similar to this:
http://www.whipplesuperchargers.com/...sp?ProdID=1196
Any other ideas? Anyone with similar problems or solutions?
I have been using the Corvette for about a week now and I have no major problems. I have done a few hours of tuning and it is running quite good now. The tranny just broke in my winter beater so the Corvette is my daily driver right now.
I wired the injectors like the stock TPI. Right side on one channel and left side on one channel, batch fire. According the my WB lambda I need to run it a bit rich at about 2500-3000 rpm. WOT timing is constant in that range. If I keep WOT lambda constant it starts to miss in that area, and then it runs fine above. I think it is because of some resonace in the fuel pressure resulting in different fuel pressure for different injectors. I am using 120 lb/hr injectors and fireing all four in the same fuel rail at the same time might be too much. I am going to re-wire the injectors to use one injector from each corner as a group. If I add fuel in that area the problem goes away and I think that it is because the 'lean' cylinders get enough fuel. The 'rich' ones are probably way rich. I also tested with another nr of spark pulses per injector pulse, that changes both the injector frequency and pulse length. When I did that the 'problem' rpm range also changed. This is the reson I suspect some resonace in the fuel pressure.
I probably need to add a fuel pressure damper too, something similar to this:
http://www.whipplesuperchargers.com/...sp?ProdID=1196
Any other ideas? Anyone with similar problems or solutions?
Last edited by JoBy; 06-20-2006 at 09:30 AM.
06-20-2006, 10:32 AM
#2
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
What ECM and what BIN are you using?
I did quite a bit of ECM bench testing of the $58 BBZB code quasi-fueling this weekend. I would see all kinds of strange things happen with it at different RPMs and BPWs. If you use the $58 code, make sure the injector BPW cal data is set that you won't enter quasi-fuel mode (it is reported in the ALDL data). If you hit quasi mode you will see strange fueling kind of like what you described.
You are doing alternate firing? Maybe an aftermarket or Megasquirt EFI?
All 165, 730, 749 ECMs fire all 8 injectors at the same time. There is no way around it unless you cut & jumper the PCB and re-write code. Therefore, it doesn't matter how you wire it with those GM ECMs
I did quite a bit of ECM bench testing of the $58 BBZB code quasi-fueling this weekend. I would see all kinds of strange things happen with it at different RPMs and BPWs. If you use the $58 code, make sure the injector BPW cal data is set that you won't enter quasi-fuel mode (it is reported in the ALDL data). If you hit quasi mode you will see strange fueling kind of like what you described.
You are doing alternate firing? Maybe an aftermarket or Megasquirt EFI?
All 165, 730, 749 ECMs fire all 8 injectors at the same time. There is no way around it unless you cut & jumper the PCB and re-write code. Therefore, it doesn't matter how you wire it with those GM ECMs
06-20-2006, 12:37 PM
#3
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
I really don't think it is a mechanical/fuel pressure problem.
I think it due to the very small BPWs. What is the BPWs when this problem occurs and what are they when it does not occur?
I think it due to the very small BPWs. What is the BPWs when this problem occurs and what are they when it does not occur?
06-20-2006, 02:47 PM
#4
Senior Member
Thread Starter
Join Date: Oct 2001
Location: Timrå, Sweden
Posts: 930
Likes: 0
Received 0 Likes
on
0 Posts
Car: 1984 Corvette
Engine: Turbo 350
Transmission: 4L80E with TCI T-Com
I am using a Haltech E6K. When I select 'batch fire' it is batch fire all the time.
At first I had a bad idle running on only a few cylinders. Pulse width were about 1ms at idle. When I loaded the engine all cylinders came to life. Then at about 3500-4000 rpm it started missing again like I described above.
A discoverd that I was using ingition/2 as injector frequency. The injectors were fireing twice for each rpm, four times on the 4-stroke cycle.
I changed to ignition/4 as injector frequency so the injectors only fired once every revolution. Now I got a steady idle using about 2ms injector time. When I did this the 'strange miss' at WOT moved from 3500-4000 rpm to about 2500-3000 rpm. Probably where the injectors had about the same pulse length.
At first I had a bad idle running on only a few cylinders. Pulse width were about 1ms at idle. When I loaded the engine all cylinders came to life. Then at about 3500-4000 rpm it started missing again like I described above.
A discoverd that I was using ingition/2 as injector frequency. The injectors were fireing twice for each rpm, four times on the 4-stroke cycle.
I changed to ignition/4 as injector frequency so the injectors only fired once every revolution. Now I got a steady idle using about 2ms injector time. When I did this the 'strange miss' at WOT moved from 3500-4000 rpm to about 2500-3000 rpm. Probably where the injectors had about the same pulse length.
Last edited by JoBy; 06-20-2006 at 02:52 PM.
06-20-2006, 03:52 PM
#5
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Something doesn't make sense with your BPW and injector firing scheme. I know with complete certainty that my high-z 42#/hr injectors fire double fire batch (all 8 injs fire twice in one engine rev. or every 4 Ign. events) I currently have BPWs of 1.1ms at about 14.7AFR at 800 RPM.
I can't see how your 120#/hr have BPWs at idle greater than mine. Maybe I am misunderstanding your firing method.
There is only two things I can think of. 1) that the ECM is in a fueling transitional mode where the miss occurs. Maybe going from one fueling mode to another. Not sure, I have no idea how the aftermarkets do it. 2) Bad injector(s)??? seems kind of far fetch, but ya never know. Did you bench them to see the spray pattern vs. BPWs?
Sorry I can't be more help. I don't know how the aftermarket ECMs work. I would guess it is a fuel transition mode. Can you see the async BPW change when the miss occurs?
With injectors that big, I would try "alternate bank" firing with it firing every ign/2 (fire 4 injectors alternating fashion every 2 ign events). This would give BPWs the same as firing all 8 injs every ign/4. Well, it gives the same amount of fuel but should be more smooth. Using either method, it delivers the same amount of fuel per 4-stroke......I still don't see how your 120#/hr want the same BPWs as my 42#/hrs. I think you are off by a factor of two somewhere or the injectors are firing scheme is different than you think.
That is commonly referred to as double fire batch mode (all 8 injectors squirt once per crank revolution or twice every engine revolution). This is the normal TPI firing mode. You would need a huge VE at idle if what you said were true. Do the quick Calc of your estimated HP and what the BPWs (or duty cycles) would be. Then use a VE of 20%-30% for your idle and see what the BPWs come out to.
I would try to log data and look for a fuel mode change done by the ECM. Does the ECM report the injector firing mode like the GM TPI ECMs do? That is, single fire, double fire, alternate fire, etc? For example, the GM $8D code reports the BPW and firing mode on the ALDL data stream. If it is in double fire it will report the BPW that was sent to the injectors. So, the real on time per 4-stroke is 2*BPW. If in single fire mode, the real on time per 4-stroke is 1*BPW. The user needs to do the math, the ECM does not do it for you. Is the aftermarket ECM the same way?? It matters.
I can't see how your 120#/hr have BPWs at idle greater than mine. Maybe I am misunderstanding your firing method.
There is only two things I can think of. 1) that the ECM is in a fueling transitional mode where the miss occurs. Maybe going from one fueling mode to another. Not sure, I have no idea how the aftermarkets do it. 2) Bad injector(s)??? seems kind of far fetch, but ya never know. Did you bench them to see the spray pattern vs. BPWs?
Sorry I can't be more help. I don't know how the aftermarket ECMs work. I would guess it is a fuel transition mode. Can you see the async BPW change when the miss occurs?
With injectors that big, I would try "alternate bank" firing with it firing every ign/2 (fire 4 injectors alternating fashion every 2 ign events). This would give BPWs the same as firing all 8 injs every ign/4. Well, it gives the same amount of fuel but should be more smooth. Using either method, it delivers the same amount of fuel per 4-stroke......I still don't see how your 120#/hr want the same BPWs as my 42#/hrs. I think you are off by a factor of two somewhere or the injectors are firing scheme is different than you think.
Originally Posted by JoBy
I changed to ignition/4 as injector frequency so the injectors only fired once every revolution. Now I got a steady idle using about 2ms injector time.
I would try to log data and look for a fuel mode change done by the ECM. Does the ECM report the injector firing mode like the GM TPI ECMs do? That is, single fire, double fire, alternate fire, etc? For example, the GM $8D code reports the BPW and firing mode on the ALDL data stream. If it is in double fire it will report the BPW that was sent to the injectors. So, the real on time per 4-stroke is 2*BPW. If in single fire mode, the real on time per 4-stroke is 1*BPW. The user needs to do the math, the ECM does not do it for you. Is the aftermarket ECM the same way?? It matters.
06-20-2006, 06:59 PM
#6
Senior Member
Thread Starter
Join Date: Oct 2001
Location: Timrå, Sweden
Posts: 930
Likes: 0
Received 0 Likes
on
0 Posts
Car: 1984 Corvette
Engine: Turbo 350
Transmission: 4L80E with TCI T-Com
This is from the Haltech E6K manual.
As I understand this the Haltech does not change injection mode. It is what you select.
My current selection is 'Batch-fire' and 'Ign / By' = 4.
At idle the injection time is about 2ms and lambda is good.
With 'Ign / By' = 4 the injectors should fire once every crank revolution.
Combined with 'Batch-fire' it might be that half the injectors fire every crank revolution. It is also possible that all injectors fire once every crank revolution. I have to bring the oscilloscope out to be sure.
I get 1ms or 2ms total injection time for every crank revolution depending on that.
If all your injectors fire twice in one crank revolution, then you have 2.2ms injection time for every crank revolution.
I don't know the exact size of the injectors but I measured them and as I measured they are in the 120# range. I am using a bit lower fuel pressure than I measured them with, but that should not make a big differance.
During WOT the injector duty cycle climbs to about 25%-30%.
4.4.1 Fuel Set-up
The Fuel Set-up works in an identical way to the Identification. Its fields are different and relate to the way the fuel is delivered to the engine. The fields in the Fuel Set-up are:
Ign / By
Ignition Divide By is the number of ignition pulses that will be counted until the next injection pulse. For almost all multipoint systems, injection should occur once per revolution and so Ignition Divide By should be set to half the number of cylinders. If the system is operating in Batch Fire or Sequential mode, or is a rotary, then a value of 1 is suggested.
Decel Cut-Off
A common fuel saving feature in original equipment computers is a fuel cut-off on deceleration. This will cut fuel delivery to the engine while coasting down hills with closed throttle. This feature can be enabled or disabled on the E6K. It is better, when first tuning, to disable this function.
Injection Mode
Depending on the ECU settings the E6K splits its four/eight injector driver outputs into two banks (see chapter 13, 13.1). INJ1 and INJ2 comprise the first bank. INJ3 and INJ4 form Bank 2 (refer to the wiring diagram at the back of this manual.) In Basic Mode, Fuel can be injected in three different modes.
Multipoint injection fires all the injectors together. This is the most common set-up and will normally be used on engines with multipoint injection manifolds (one injector per cylinder).
Batch-fire injection is usually used in throttle body or non-turbo rotary set-ups and fires the two banks of injectors alternately. On eight and twelve injector fuel rails, with high-flow injectors, this may also help reduce fuel pressure oscillations caused by all injectors pulsing together.
Staged injection is usually used on high boost turbo engines. Injector Bank 1 fires all the time, just as in a multipoint set-up. Beyond a set boost pressure, the second bank of injectors is enabled. These "staged" injectors are normally upstream of the primary injectors, adding to their fuel flow. The point at which the E6K switches in the secondary injectors is set via the Staging Bar Number field. Staging permits high fuelflow capability, but maintains accuracy and controllability at light load and idle. See Appendix D [D.3] for more details on staging. In Advanced Mode, Sequential Injection can also be selected. This option is not a straight-forward set-up. It requires more outputs for fuel than normally used. Before selecting this option carefully read the section on sequential injection in Appendix B -
The Fuel Set-up works in an identical way to the Identification. Its fields are different and relate to the way the fuel is delivered to the engine. The fields in the Fuel Set-up are:
Ign / By
Ignition Divide By is the number of ignition pulses that will be counted until the next injection pulse. For almost all multipoint systems, injection should occur once per revolution and so Ignition Divide By should be set to half the number of cylinders. If the system is operating in Batch Fire or Sequential mode, or is a rotary, then a value of 1 is suggested.
Decel Cut-Off
A common fuel saving feature in original equipment computers is a fuel cut-off on deceleration. This will cut fuel delivery to the engine while coasting down hills with closed throttle. This feature can be enabled or disabled on the E6K. It is better, when first tuning, to disable this function.
Injection Mode
Depending on the ECU settings the E6K splits its four/eight injector driver outputs into two banks (see chapter 13, 13.1). INJ1 and INJ2 comprise the first bank. INJ3 and INJ4 form Bank 2 (refer to the wiring diagram at the back of this manual.) In Basic Mode, Fuel can be injected in three different modes.
Multipoint injection fires all the injectors together. This is the most common set-up and will normally be used on engines with multipoint injection manifolds (one injector per cylinder).
Batch-fire injection is usually used in throttle body or non-turbo rotary set-ups and fires the two banks of injectors alternately. On eight and twelve injector fuel rails, with high-flow injectors, this may also help reduce fuel pressure oscillations caused by all injectors pulsing together.
Staged injection is usually used on high boost turbo engines. Injector Bank 1 fires all the time, just as in a multipoint set-up. Beyond a set boost pressure, the second bank of injectors is enabled. These "staged" injectors are normally upstream of the primary injectors, adding to their fuel flow. The point at which the E6K switches in the secondary injectors is set via the Staging Bar Number field. Staging permits high fuelflow capability, but maintains accuracy and controllability at light load and idle. See Appendix D [D.3] for more details on staging. In Advanced Mode, Sequential Injection can also be selected. This option is not a straight-forward set-up. It requires more outputs for fuel than normally used. Before selecting this option carefully read the section on sequential injection in Appendix B -
My current selection is 'Batch-fire' and 'Ign / By' = 4.
At idle the injection time is about 2ms and lambda is good.
With 'Ign / By' = 4 the injectors should fire once every crank revolution.
Combined with 'Batch-fire' it might be that half the injectors fire every crank revolution. It is also possible that all injectors fire once every crank revolution. I have to bring the oscilloscope out to be sure.
I get 1ms or 2ms total injection time for every crank revolution depending on that.
Originally Posted by junkcltr
Something doesn't make sense with your BPW and injector firing scheme. I know with complete certainty that my high-z 42#/hr injectors fire double fire batch (all 8 injs fire twice in one engine rev. or every 4 Ign. events) I currently have BPWs of 1.1ms at about 14.7AFR at 800 RPM.
I can't see how your 120#/hr have BPWs at idle greater than mine. Maybe I am misunderstanding your firing method.
I can't see how your 120#/hr have BPWs at idle greater than mine. Maybe I am misunderstanding your firing method.
I don't know the exact size of the injectors but I measured them and as I measured they are in the 120# range. I am using a bit lower fuel pressure than I measured them with, but that should not make a big differance.
During WOT the injector duty cycle climbs to about 25%-30%.
06-20-2006, 07:23 PM
#7
Senior Member
Thread Starter
Join Date: Oct 2001
Location: Timrå, Sweden
Posts: 930
Likes: 0
Received 0 Likes
on
0 Posts
Car: 1984 Corvette
Engine: Turbo 350
Transmission: 4L80E with TCI T-Com
Also from the Haltech E6K manual.
D.5 Fuel Rails and Pressure Regulators
A long fuel rail with narrow internal diameter will suffer from pulsation in the fuel rail. The internal rail diameter should be around 12mm (œ"). Even so, oscillations may occur, particularly if the injectors are large. A fuel damper can help in removing these oscillations. If running a multipoint set-up, batch fire injection will also reduce oscillation amplitude. Oscillations may occur only within a certain rpm range, so a fuel pressure meter should be monitored throughout the driving range of the engine. On V configuration motors, it may be more practical to employ two regulators, one on each bank's rail. The fuel from the pump can be split to the two rails, and the return line from the
two regulators can be joined. Since pressure regulators work within certain flow limits, this may also save having to purchase an expensive high pressure / high flow regulator.
A long fuel rail with narrow internal diameter will suffer from pulsation in the fuel rail. The internal rail diameter should be around 12mm (œ"). Even so, oscillations may occur, particularly if the injectors are large. A fuel damper can help in removing these oscillations. If running a multipoint set-up, batch fire injection will also reduce oscillation amplitude. Oscillations may occur only within a certain rpm range, so a fuel pressure meter should be monitored throughout the driving range of the engine. On V configuration motors, it may be more practical to employ two regulators, one on each bank's rail. The fuel from the pump can be split to the two rails, and the return line from the
two regulators can be joined. Since pressure regulators work within certain flow limits, this may also save having to purchase an expensive high pressure / high flow regulator.
Trending Topics
06-20-2006, 11:52 PM
#8
Supreme Member
iTrader: (1)
Join Date: Jan 2002
Location: garage
Posts: 4,432
Likes: 0
Received 1 Like
on
1 Post
Engine: 3xx ci tubo
Transmission: 4L60E & 4L80E
Originally Posted by JoBy
As I understand this the Haltech does not change injection mode. It is what you select.
My current selection is 'Batch-fire' and 'Ign / By' = 4.
At idle the injection time is about 2ms and lambda is good.
With 'Ign / By' = 4 the injectors should fire once every crank revolution.
Combined with 'Batch-fire' it might be that half the injectors fire every crank revolution. It is also possible that all injectors fire once every crank revolution. I have to bring the oscilloscope out to be sure.
My current selection is 'Batch-fire' and 'Ign / By' = 4.
At idle the injection time is about 2ms and lambda is good.
With 'Ign / By' = 4 the injectors should fire once every crank revolution.
Combined with 'Batch-fire' it might be that half the injectors fire every crank revolution. It is also possible that all injectors fire once every crank revolution. I have to bring the oscilloscope out to be sure.
That is, it squirts 4 injectors in one bank in the first crank rev., then 4 injectors in the other bank of the crank rev. during a 4-stroke revolution. If their docs are correct, then you don't really need the O-scope.
Originally Posted by JoBy
I get 1ms or 2ms total injection time for every crank revolution depending on that.
If all your injectors fire twice in one crank revolution, then you have 2.2ms injection time for every crank revolution.
If all your injectors fire twice in one crank revolution, then you have 2.2ms injection time for every crank revolution.
in one 4-stroke cycle with 120#/hr injectors. You are squirting roughly half the fuel in terms of time, but 2.5-3x in terms of flow. Or about 3/2 the fuel that mine setup squirts. That sounds reasonable.
Originally Posted by JoBy
I don't know the exact size of the injectors but I measured them and as I measured they are in the 120# range. I am using a bit lower fuel pressure than I measured them with, but that should not make a big differance.
During WOT the injector duty cycle climbs to about 25%-30%.
During WOT the injector duty cycle climbs to about 25%-30%.
Can you see the miss show up in the AFR log? I know when I had spark blow-out with a 10 AFR at WOT I could see it in the ALDL data from the WBO2. But that was with more cyls missing.
With the small BPWs and the 120#/hr injectors I can't see the pulsation/oscillation really occuring in the fuel rails because there isn't a large amount of fuel being delivered to the engine. I can't see your 30% DCs with 120#/hr any different than my 80% DCs with 42#/hr injectors.
I could be wrong though. My first instinct was an injector not flowing like the others and more cyl pressure (w/o boost even) causing the miss due to less fuel. I would richen it up at that spot and see if the miss goes away.
I was thinking that the 2500-3000 RPM lean area was due to the cam coming in, but it sounds like the miss moved when you changed the inject scheme. Not sure what to tell you.
06-21-2006, 03:23 AM
#9
Senior Member
Thread Starter
Join Date: Oct 2001
Location: Timrå, Sweden
Posts: 930
Likes: 0
Received 0 Likes
on
0 Posts
Car: 1984 Corvette
Engine: Turbo 350
Transmission: 4L80E with TCI T-Com
Originally Posted by haltech
A fuel damper can help in removing these oscillations. If running a multipoint set-up, batch fire injection will also reduce oscillation amplitude. Oscillations may occur only within a certain rpm range
Originally Posted by whipplesuperchargers
A damper is critical to the proper function of any EFI setup. Rapid cycling of the fuel injectors and fuel system components in a running engine combine with the non-compressibility of the fuel; and create high frequency pressure waves in the rail. These pressure waves can cause substantial variation in the fuel delivered by the injectors; causing numerous running problems that are difficult to trace, in addition to increasing stress on the fuel pressure regulator.
Originally Posted by Marren Fuel Injection
Batch-fire fuel systems, systems with three or more injectors on each fuel rail, or fuel systems that use high-flow injectors are especially susceptible to hydraulic pulsation, so it is especially important to use dampers on these systems.
Thread
Thread Starter
Forum
Replies
Last Post
LT1Formula
Engine/Drivetrain/Suspension Parts for Sale
7
10-08-2015 08:34 PM
3.8TransAM
NW Indiana and South Chicago Suburb
2
10-01-2015 07:47 PM
