Just a goody I wanted to share that I recently found about sequential fuel...
http://msextra.com/doc/ms3/tunerstudio.html
when you switch to sequential it only does 1 squirt per cycle.. good to know I thought wrong on this
http://msextra.com/doc/ms3/Engine_Seque ... tings.html
Sequential fuel timing
- volksbugly
- Posts: 634
- Joined: Thu Sep 16, 2010 5:09 am
- Piledriver
- Moderator
- Posts: 22518
- Joined: Sat Feb 16, 2002 12:01 am
Re: Sequential fuel timing
Those are MS3 docs, but the same is true for MS2-extra sequential until recently.
The latest alpha should have changes to allow multi-squirts ala semi sequential IIRC...
I asked for it to do some back-to-back tests (full vs. semi sequential), and one of the devs made some changes to allow it, but a MS3 has distracted me since, haven't been able to test it.
Full sequential will help your min. PW area near idle, other than that it's really just a question of if it makes a real difference elsewhere, as it was hard to do a valid comparison test.
If you are set up for sequential there is no known downside, I was just curious and also wanted to do semi-sequential/staged on the turbo setup, but it will be getting an MS3, full sequential+staged, and dual spark if I can pull it off, the rotary trailing spark code may be tweakable for trailing spark on a /4.
(I won't be coding, but the devs seemed interested, I have to decide if it's worth it to drill more holes in my heads, what I have in mind is not the normal setup, but it might make Larry Widmer smile-- one 10mm plug on each end of the wedge)
Personally I do start of squirt timing just as my intake closes, and give it 20 degrees of advance over full MAP and 40 with RPM. These numbers were pulled entirely from thin air, but seem to work for me, YMMV.
There should be a way of calculating the delay and proper advance time to the valve if the injected fuel flight speed vs pressure and distance are known, but ...
The latest alpha should have changes to allow multi-squirts ala semi sequential IIRC...
I asked for it to do some back-to-back tests (full vs. semi sequential), and one of the devs made some changes to allow it, but a MS3 has distracted me since, haven't been able to test it.
Full sequential will help your min. PW area near idle, other than that it's really just a question of if it makes a real difference elsewhere, as it was hard to do a valid comparison test.
If you are set up for sequential there is no known downside, I was just curious and also wanted to do semi-sequential/staged on the turbo setup, but it will be getting an MS3, full sequential+staged, and dual spark if I can pull it off, the rotary trailing spark code may be tweakable for trailing spark on a /4.
(I won't be coding, but the devs seemed interested, I have to decide if it's worth it to drill more holes in my heads, what I have in mind is not the normal setup, but it might make Larry Widmer smile-- one 10mm plug on each end of the wedge)
Personally I do start of squirt timing just as my intake closes, and give it 20 degrees of advance over full MAP and 40 with RPM. These numbers were pulled entirely from thin air, but seem to work for me, YMMV.
There should be a way of calculating the delay and proper advance time to the valve if the injected fuel flight speed vs pressure and distance are known, but ...
Addendum to Newtons first law:
zero vehicles on jackstands, square gets a fresh 090 and 1911, cabby gets a blower.
EZ3.6 Vanagon after that.(mounted, needs everything finished) then Creamsicle.
zero vehicles on jackstands, square gets a fresh 090 and 1911, cabby gets a blower.
EZ3.6 Vanagon after that.(mounted, needs everything finished) then Creamsicle.
- Piledriver
- Moderator
- Posts: 22518
- Joined: Sat Feb 16, 2002 12:01 am
Re: Sequential fuel timing
Addendum to Newtons first law:
zero vehicles on jackstands, square gets a fresh 090 and 1911, cabby gets a blower.
EZ3.6 Vanagon after that.(mounted, needs everything finished) then Creamsicle.
zero vehicles on jackstands, square gets a fresh 090 and 1911, cabby gets a blower.
EZ3.6 Vanagon after that.(mounted, needs everything finished) then Creamsicle.
-
- Posts: 7400
- Joined: Sat Mar 10, 2001 12:01 am
Re: Sequential fuel timing
Pyle,
I have a lot of the "Widmer" that you speak of. Anti reversion infinite angle exhaust seat, squishies, etc. The Redline ecu has eight separate ignition drivers with completely adjustable timing per cylinder. I want to go dual plug just so I can play with it.
Pyle, your maps on the previous page have a big change in AFR targets at 2300. Does it hunt or act strange there? It goes from low 14s to mid 16s in 100 rpms. I understand why you make the fast change, but what happens when you are right on the hump with the drivability?
I have a lot of the "Widmer" that you speak of. Anti reversion infinite angle exhaust seat, squishies, etc. The Redline ecu has eight separate ignition drivers with completely adjustable timing per cylinder. I want to go dual plug just so I can play with it.
Pyle, your maps on the previous page have a big change in AFR targets at 2300. Does it hunt or act strange there? It goes from low 14s to mid 16s in 100 rpms. I understand why you make the fast change, but what happens when you are right on the hump with the drivability?
Steve
My Baja Build
My Baja Build
- Piledriver
- Moderator
- Posts: 22518
- Joined: Sat Feb 16, 2002 12:01 am
Re: Sequential fuel timing
I should say "anyone running MS2Extra v3"
(I think semi-sequential's been in there for most of v3.x since the early alphas)
You SHOULD be running at LEAST 3.21, some ugly bugs were squashed, and 3.3x fixes many more and adds a herd of idle improvements (backports from ms3) should hit gold rsn, it's quite stable IMHO.
Advanced>Sequential Injection>set to sequential/semi sequential and >injector drivers select "standard drivers"
(so you tell it to use sequential with the std drivers, and it fakes it as best it can. Try 2 squirts simultaneous. And the Veal)
TRY THE "AVERAGE" MAP sampling option. Big win. Don't look back.
Steve, it does ~nothing there, it just won't "pull" the high AFR# at lower RPM, and that's the latest spot I picked.
I'm shooting for either very lean>high timing (to provide proper advance for the slower burn) or normal/normal, and minimal interpolation, as there be dragons in the transition areas.
If I could get it to run everywhere, well, at ~20:1 (except >~87KPA MAP) I would, as it utterly headshots the CHT at cruise.
Your setup might be able to pull it off better.
The latest VE targets are running 20:1 >2600 and under 83KPA., 13:1 over 87, and 14:1:1 around idle.(~stoich on E10)
MS3 appears to have "free" timing adjustment on all 8 as well, but the proper way to do it is borrow the rotary trailing spark code, as it exists, that way you can set your base map and one offset. (although I imagine you will still be able to tweak some on each cylinder/plug if needed)
I'm running LS2 coils and Pulstar Iridiums... Much to my surprise, they do seem to make at least a small difference in the AFR# I can reliably pull w/o lean surge. (I wanted to try them, I didn't really expect much)
With the LS2 truck coils and the Pulstars, it sounds like a string of small firecrackers going off on the bench.
(I think semi-sequential's been in there for most of v3.x since the early alphas)
You SHOULD be running at LEAST 3.21, some ugly bugs were squashed, and 3.3x fixes many more and adds a herd of idle improvements (backports from ms3) should hit gold rsn, it's quite stable IMHO.
Advanced>Sequential Injection>set to sequential/semi sequential and >injector drivers select "standard drivers"
(so you tell it to use sequential with the std drivers, and it fakes it as best it can. Try 2 squirts simultaneous. And the Veal)
TRY THE "AVERAGE" MAP sampling option. Big win. Don't look back.
Steve, it does ~nothing there, it just won't "pull" the high AFR# at lower RPM, and that's the latest spot I picked.
I'm shooting for either very lean>high timing (to provide proper advance for the slower burn) or normal/normal, and minimal interpolation, as there be dragons in the transition areas.
If I could get it to run everywhere, well, at ~20:1 (except >~87KPA MAP) I would, as it utterly headshots the CHT at cruise.
Your setup might be able to pull it off better.
The latest VE targets are running 20:1 >2600 and under 83KPA., 13:1 over 87, and 14:1:1 around idle.(~stoich on E10)
MS3 appears to have "free" timing adjustment on all 8 as well, but the proper way to do it is borrow the rotary trailing spark code, as it exists, that way you can set your base map and one offset. (although I imagine you will still be able to tweak some on each cylinder/plug if needed)
I'm running LS2 coils and Pulstar Iridiums... Much to my surprise, they do seem to make at least a small difference in the AFR# I can reliably pull w/o lean surge. (I wanted to try them, I didn't really expect much)
With the LS2 truck coils and the Pulstars, it sounds like a string of small firecrackers going off on the bench.
Addendum to Newtons first law:
zero vehicles on jackstands, square gets a fresh 090 and 1911, cabby gets a blower.
EZ3.6 Vanagon after that.(mounted, needs everything finished) then Creamsicle.
zero vehicles on jackstands, square gets a fresh 090 and 1911, cabby gets a blower.
EZ3.6 Vanagon after that.(mounted, needs everything finished) then Creamsicle.
- raygreenwood
- Posts: 11895
- Joined: Wed Jan 22, 2003 12:01 am
Re: Sequential fuel timing
Coming in late to this discussion...but I have been experimenting with sequential for quite a few years mainly on stock injection systems (D-jet, and CIS). Reference toward Piledrivers earlier statements and somehwat to Steve Arndt's.....some of the actual pro's and cons I have found.
1) injecting at valve closed versus open: Injecting at valve closed may in fact help fuel vaporize better...but it can be a tradeoff. Depending on injector placement, getting fuel spray all over the port walls actually makes fuel feeding worse because it takes roughly 5X the airflow to remove fuel wetted to port walls that it does to just pull it in with the airstream. This is also why on sequential systems and on injection in general...smooth port walls are better than rough ones. Its far easier to remove wetted fuel from polished or smooth cast walls than it is from textured walls. This has the habit of leaving fuel from one dosage to be used on a succeeding dosage. Thats fuel left out of one injection cycle...and used on a later one. Conventional wisdom might lead you to think that this will eventually even out....but depending on how you are driving at the time.....it doesn't really work that cleanly. This is easily seen in the D-jet systems by the two filthy intake ports they typically have from the two out of phase injections. L-jet ports and most banked injection are also generally filthy (with lots of city driving).
2.) spray pattern and pattern width: As pile noted...spraying at he back of the valve is critical. That means a spray pattern just the width of the valve and seat to keep the fuel off the port walls is ideal. If you can hit the hot valve a couple of milliseconds before it opens....that helps with vaporization.
The real issue is atomization here. Injectors do a poor job of atomization as compared to carbs. I have found the best way is to get the fuel atomized as best as possible at the injector 9some pintles and tip shapes do better than others)....but actual airflow and turbulence do the best work. Its very hard to know on any given engine when and where the actual best turbulence or shear causing airflow is.
One thing I have found is that ...at least on my engines....it is the least beneficial right at the point where the intake valve is lifting off the seat. At low rpms at worst the airflow in that area is static for a few milliseconds. At higher rpms....there is plenty of turbulence from reversion at the moment the valve lifts from the seat...but its everywhere but where you want the fuel to be. Usually starting at about 1/4-1/3 lift depending on cam and valve size ....is where beneficial velocity to draw just injected fuel into a fast moving airstream seems to start.
Also if you have huge valves....its not quite as beneficial because just like using a TB taht is too large will drop airflow velocity at the TB plate with small throttle movements...opening an overly large intake valve dumps airflow velocity very quickly.
Using sequential seems to have its best benefits when velocities at the valve stay high. this is about velocity...not volume.
3.) Timing and advance of injection: From the above i found that...as noted...when you start the injection in sequential is critical...and different for every combo. But when you shut it off is important...though not critical......but is part of the planning for what size injectors you will need. You will need to advance your sequential injection events as rpm rises....or you will not be able to complete the required injection dosage before the valve slams shut. If that happens.....run lean and the purpose of all of this is defeated. This means that the injector size must be carefully considered. Get one too large and you can't control it at low rpms. get one too small and you cannot dleiver enough fuel sequentially to complete the required dosage.
Fuel pressure to leverage this is critical.
I found that for the most part that much over 3000 rpm....sequential does not really help that much at all. The time interval is to short. At sustained medium to high rpm driving.....with your injection events advanced out to injected marginally before the valve opens....but close enough that fuel spray never has a millisecond to alight on the port walls....the time interval is too short between valve open and valve closed to make much difference. But when its working...throttle resposne will be much much better.
4.) bits and pieces: this is all about atomization as noted earlier. If through all of your efforts...timing, sequential injection etc......you are not greatly improviong fuel atomization....you will not see much more than maybe a slight improvement in AFR at the tailpipe nad maybe slightly better economy at certain speeds.
Anything you can do improve atomization right before valve lift will help you. In fact if your initial injector atomization and placement are very poor.....sequential injection may show you no benefit at all.
We have both benefits and issues to contend with when using the stock type 3 and 4 runner tubes. The benefits are fairly smooth flow and pretty damned good velocity at the lower runner port. One of the issues though is that when the airflow rounds the outside bend.....its force is slung to the outside wall. This airflow tends to force/slam the injector spray pattern down onto the lower port wall. Injector aim will be critical
The distance on stock type 4 heads for instance is fairly long between the injector and the back of the valve. I am finding that some of the newer higher pressure injectors have too wide of a spray cone for this distance. More fuel on the port walls and less in the port.
I am experimenting with some machined aluminum bells or cones that i took off of some injectors about 10 years ago....a volvo turbo maybe? They look like rocket nozzles but their function was to keep the spray pattern in a tighter cone....and extend it closer to the intake valve. I am also work on some pieces that i can only really describe as augmentors or diffusers for injector tips that should create a much finer spray pattern.
Sorry for the length.....but I mainly wanted to relay.....that sequential injection is not just about timing and fuel maps. Its turbulence, airflow etc. Ray
1) injecting at valve closed versus open: Injecting at valve closed may in fact help fuel vaporize better...but it can be a tradeoff. Depending on injector placement, getting fuel spray all over the port walls actually makes fuel feeding worse because it takes roughly 5X the airflow to remove fuel wetted to port walls that it does to just pull it in with the airstream. This is also why on sequential systems and on injection in general...smooth port walls are better than rough ones. Its far easier to remove wetted fuel from polished or smooth cast walls than it is from textured walls. This has the habit of leaving fuel from one dosage to be used on a succeeding dosage. Thats fuel left out of one injection cycle...and used on a later one. Conventional wisdom might lead you to think that this will eventually even out....but depending on how you are driving at the time.....it doesn't really work that cleanly. This is easily seen in the D-jet systems by the two filthy intake ports they typically have from the two out of phase injections. L-jet ports and most banked injection are also generally filthy (with lots of city driving).
2.) spray pattern and pattern width: As pile noted...spraying at he back of the valve is critical. That means a spray pattern just the width of the valve and seat to keep the fuel off the port walls is ideal. If you can hit the hot valve a couple of milliseconds before it opens....that helps with vaporization.
The real issue is atomization here. Injectors do a poor job of atomization as compared to carbs. I have found the best way is to get the fuel atomized as best as possible at the injector 9some pintles and tip shapes do better than others)....but actual airflow and turbulence do the best work. Its very hard to know on any given engine when and where the actual best turbulence or shear causing airflow is.
One thing I have found is that ...at least on my engines....it is the least beneficial right at the point where the intake valve is lifting off the seat. At low rpms at worst the airflow in that area is static for a few milliseconds. At higher rpms....there is plenty of turbulence from reversion at the moment the valve lifts from the seat...but its everywhere but where you want the fuel to be. Usually starting at about 1/4-1/3 lift depending on cam and valve size ....is where beneficial velocity to draw just injected fuel into a fast moving airstream seems to start.
Also if you have huge valves....its not quite as beneficial because just like using a TB taht is too large will drop airflow velocity at the TB plate with small throttle movements...opening an overly large intake valve dumps airflow velocity very quickly.
Using sequential seems to have its best benefits when velocities at the valve stay high. this is about velocity...not volume.
3.) Timing and advance of injection: From the above i found that...as noted...when you start the injection in sequential is critical...and different for every combo. But when you shut it off is important...though not critical......but is part of the planning for what size injectors you will need. You will need to advance your sequential injection events as rpm rises....or you will not be able to complete the required injection dosage before the valve slams shut. If that happens.....run lean and the purpose of all of this is defeated. This means that the injector size must be carefully considered. Get one too large and you can't control it at low rpms. get one too small and you cannot dleiver enough fuel sequentially to complete the required dosage.
Fuel pressure to leverage this is critical.
I found that for the most part that much over 3000 rpm....sequential does not really help that much at all. The time interval is to short. At sustained medium to high rpm driving.....with your injection events advanced out to injected marginally before the valve opens....but close enough that fuel spray never has a millisecond to alight on the port walls....the time interval is too short between valve open and valve closed to make much difference. But when its working...throttle resposne will be much much better.
4.) bits and pieces: this is all about atomization as noted earlier. If through all of your efforts...timing, sequential injection etc......you are not greatly improviong fuel atomization....you will not see much more than maybe a slight improvement in AFR at the tailpipe nad maybe slightly better economy at certain speeds.
Anything you can do improve atomization right before valve lift will help you. In fact if your initial injector atomization and placement are very poor.....sequential injection may show you no benefit at all.
We have both benefits and issues to contend with when using the stock type 3 and 4 runner tubes. The benefits are fairly smooth flow and pretty damned good velocity at the lower runner port. One of the issues though is that when the airflow rounds the outside bend.....its force is slung to the outside wall. This airflow tends to force/slam the injector spray pattern down onto the lower port wall. Injector aim will be critical
The distance on stock type 4 heads for instance is fairly long between the injector and the back of the valve. I am finding that some of the newer higher pressure injectors have too wide of a spray cone for this distance. More fuel on the port walls and less in the port.
I am experimenting with some machined aluminum bells or cones that i took off of some injectors about 10 years ago....a volvo turbo maybe? They look like rocket nozzles but their function was to keep the spray pattern in a tighter cone....and extend it closer to the intake valve. I am also work on some pieces that i can only really describe as augmentors or diffusers for injector tips that should create a much finer spray pattern.
Sorry for the length.....but I mainly wanted to relay.....that sequential injection is not just about timing and fuel maps. Its turbulence, airflow etc. Ray