Increasing injector flow rate - remap needed?
Discussion
Need more advice from clever people. I tried talking to myself, but as I'm not clever it wasn't useful.
We have a running race car, with a fully programmable ECU. However, the injector duty cycle looks like it's reaching 100% at high revs WOT. Given that air temp or coolant temp correction might be increasing pulse width beyond the base map, and accel enrichment might also be adding to it, it seems as though I'd be better off with bigger injectors.
We currently use Bosch injectors, rated at 218cc/min at 3.0 bar, but we actually operate at 3.5 bar (so our actual flow is going to be closer to 235.5cc/minute.
To get from 100% duty cycle to 80% duty cycle means, in theory, we need 25% more flow (100/80). Therefore it would seem that we need to either have 294cc/min at 3 bar (and run them at 3 bar) or 275cc/min at 3 bar (but remain at 3.5 bar). All well and good, I think.
However, if I did this, obviously it would inject far too much fuel unless I modified the map. Given that ~1ms of the pulse width is the opening time (very little flow for that period), would it be possible to simply reduce the pulse widths by a constant percentage to maintain the correct running/AFR/power, or will it not work properly without a full remap?
It would be nice if, say, I had an 18ms pulsewidth. 17ms of which is useful. 17/1.25 = 13.6. 13.6ms useful, plus the 1ms equals a final PW of 14.7. Would that inject the same amount of fuel? Or is the relationship totally non-linear?
Thanks in advance.
We have a running race car, with a fully programmable ECU. However, the injector duty cycle looks like it's reaching 100% at high revs WOT. Given that air temp or coolant temp correction might be increasing pulse width beyond the base map, and accel enrichment might also be adding to it, it seems as though I'd be better off with bigger injectors.
We currently use Bosch injectors, rated at 218cc/min at 3.0 bar, but we actually operate at 3.5 bar (so our actual flow is going to be closer to 235.5cc/minute.
To get from 100% duty cycle to 80% duty cycle means, in theory, we need 25% more flow (100/80). Therefore it would seem that we need to either have 294cc/min at 3 bar (and run them at 3 bar) or 275cc/min at 3 bar (but remain at 3.5 bar). All well and good, I think.
However, if I did this, obviously it would inject far too much fuel unless I modified the map. Given that ~1ms of the pulse width is the opening time (very little flow for that period), would it be possible to simply reduce the pulse widths by a constant percentage to maintain the correct running/AFR/power, or will it not work properly without a full remap?
It would be nice if, say, I had an 18ms pulsewidth. 17ms of which is useful. 17/1.25 = 13.6. 13.6ms useful, plus the 1ms equals a final PW of 14.7. Would that inject the same amount of fuel? Or is the relationship totally non-linear?
Thanks in advance.
Edited by tristancliffe on Wednesday 5th January 11:52
Assuming you have an accurate AFR guage in the car a "gross" linear correction will get you into the right sort of ball park. Most aftermarket EMS systems have an "overall" trim value to the final pulse width that could be used to determine the magnitude of the change required.
(i.e. run car at a repeatable operating point on original injectors, note AFR, swap to new injectors, re run operating point (ensuring engine and air temp corrections etc are similar) then adjust overall trim to restore lambda value.)
generally most injectors these days are relatively linear once out of the low pulse width operating slope, so doing this should be ok, but you a) might find your idle / trailing throttle AFRs are a bit wonky, and b) you deffinately want to check your WOT and higher load zones for correct AFR before using the car in anger!
(i.e. run car at a repeatable operating point on original injectors, note AFR, swap to new injectors, re run operating point (ensuring engine and air temp corrections etc are similar) then adjust overall trim to restore lambda value.)
generally most injectors these days are relatively linear once out of the low pulse width operating slope, so doing this should be ok, but you a) might find your idle / trailing throttle AFRs are a bit wonky, and b) you deffinately want to check your WOT and higher load zones for correct AFR before using the car in anger!
Edited by anonymous-user on Wednesday 5th January 13:28
Have a look in your EMS software. I know Motec has the ability for the user to adjust the injector pulse width that 100 on the main table refers to. i.e a figure of 10 equals 10m/s at 100 in the main fuel table, 12 would equal 12m/s @100 in the main fuel table. This allows the user to adjust the whole fuel map up or down.
Is there a setting that refers to something similar in your software.
Steve
Is there a setting that refers to something similar in your software.
Steve
Thanks Max, good answer.
I appreciate that it would be a bit of a guess, but your way of determining it sounds sensible and plausible. As we log AFR (and can show it on the dash), I can ensure the idle PW correction gives correct mixtures, and also ensure that the engine is safe on track - i.e. trickle back to the pits if it's dangerously lean (or indeed rich).
The advice from the ECU people was to try it, and they've recommended some VXR injectors which are rated at 440cc/min at 3 bar - 171% more than we're running at the moment, so the injector times would need to be about 62% of the current ones. This will have the added advantage of taking load off the fuel pump, which will conserve the battery a bit more - it's always a bit touch and go on voltage! But the injectors are available for £90 for a set (plus fuel rail and regulator, which I don't need), which is pretty much the same price as most injectors are individually.
But it's an experiment, and it seems not too stupid!!!
Edit: There isn't an option to scale it all in one go like that. The software allows the whole map (or parts of it) to be modified with multiply/divide/add/subtract, so I can change the whole map by 62% if I wanted to. However, the better way would be to use a spreadsheet and calculate it that way, as I can more easily factor in the opening times of the injector - hence 64% near idle and 60% at high revs. The EMS software just shows base pulsewidths in ms.
I appreciate that it would be a bit of a guess, but your way of determining it sounds sensible and plausible. As we log AFR (and can show it on the dash), I can ensure the idle PW correction gives correct mixtures, and also ensure that the engine is safe on track - i.e. trickle back to the pits if it's dangerously lean (or indeed rich).
The advice from the ECU people was to try it, and they've recommended some VXR injectors which are rated at 440cc/min at 3 bar - 171% more than we're running at the moment, so the injector times would need to be about 62% of the current ones. This will have the added advantage of taking load off the fuel pump, which will conserve the battery a bit more - it's always a bit touch and go on voltage! But the injectors are available for £90 for a set (plus fuel rail and regulator, which I don't need), which is pretty much the same price as most injectors are individually.
But it's an experiment, and it seems not too stupid!!!
Edit: There isn't an option to scale it all in one go like that. The software allows the whole map (or parts of it) to be modified with multiply/divide/add/subtract, so I can change the whole map by 62% if I wanted to. However, the better way would be to use a spreadsheet and calculate it that way, as I can more easily factor in the opening times of the injector - hence 64% near idle and 60% at high revs. The EMS software just shows base pulsewidths in ms.
Edited by tristancliffe on Wednesday 5th January 16:16
[quote=tristancliffe]Thanks Max, good answer.
I appreciate that it would be a bit of a guess, but your way of determining it sounds sensible and plausible. As we log AFR (and can show it on the dash), I can ensure the idle PW correction gives correct mixtures, and also ensure that the engine is safe on track - i.e. trickle back to the pits if it's dangerously lean (or indeed rich).
The advice from the ECU people was to try it, and they've recommended some VXR injectors which are rated at 440cc/min at 3 bar - 171% more than we're running at the moment, so the injector times would need to be about 62% of the current ones. This will have the added advantage of taking load off the fuel pump, which will conserve the battery a bit more - it's always a bit touch and go on voltage! But the injectors are available for £90 for a set (plus fuel rail and regulator, which I don't need), which is pretty much the same price as most injectors are individually.
But it's an experiment, and it seems not too stupid!!!
Edit: There isn't an option to scale it all in one go like that. The software allows the whole map (or parts of it) to be modified with multiply/divide/add/subtract, so I can change the whole map by 62% if I wanted to. However, the better way would be to use a spreadsheet and calculate it that way, as I can more easily factor in the opening times of the injector - hence 64% near idle and 60% at high revs. The EMS software just shows base pulsewidths in ms.
Edit2: By the way, I'd rather have injectors around 300cc/minute, but finding a source of particular injectors isn't easy. Anyone know of any around that for a sensible price?
I appreciate that it would be a bit of a guess, but your way of determining it sounds sensible and plausible. As we log AFR (and can show it on the dash), I can ensure the idle PW correction gives correct mixtures, and also ensure that the engine is safe on track - i.e. trickle back to the pits if it's dangerously lean (or indeed rich).
The advice from the ECU people was to try it, and they've recommended some VXR injectors which are rated at 440cc/min at 3 bar - 171% more than we're running at the moment, so the injector times would need to be about 62% of the current ones. This will have the added advantage of taking load off the fuel pump, which will conserve the battery a bit more - it's always a bit touch and go on voltage! But the injectors are available for £90 for a set (plus fuel rail and regulator, which I don't need), which is pretty much the same price as most injectors are individually.
But it's an experiment, and it seems not too stupid!!!
Edit: There isn't an option to scale it all in one go like that. The software allows the whole map (or parts of it) to be modified with multiply/divide/add/subtract, so I can change the whole map by 62% if I wanted to. However, the better way would be to use a spreadsheet and calculate it that way, as I can more easily factor in the opening times of the injector - hence 64% near idle and 60% at high revs. The EMS software just shows base pulsewidths in ms.
Edit2: By the way, I'd rather have injectors around 300cc/minute, but finding a source of particular injectors isn't easy. Anyone know of any around that for a sensible price?
Top feed, high impedance Bosch injectors as used on K-Series 1.8 engines. I believe the fitment is called EV6, although I'm not clued up enough on injector jargon to know if that is meaningful.
Sorry Max, I didn't quite follow. Are you saying that I would also have to adjust the PWs to account for the lower pressure, as in sqrt(3.5/3.0) or is there some other phenomenon you're referring to. If the pump only has to maintain 3 bar then I would hope it would draw less current/load than at 3.5 bar. But that is only a minor aspect to this really - any increase in battery life would be happily accepted.
Sorry Max, I didn't quite follow. Are you saying that I would also have to adjust the PWs to account for the lower pressure, as in sqrt(3.5/3.0) or is there some other phenomenon you're referring to. If the pump only has to maintain 3 bar then I would hope it would draw less current/load than at 3.5 bar. But that is only a minor aspect to this really - any increase in battery life would be happily accepted.
But there is a big difference between an "remap" that generally means tailoring the injection quantity to the engines volumetric efficiency curve, and a "tweak" that is a simple change of injector size. The former, needs a lot of time and some rollers / dyno to do properly, but if you already have the basic mapping done, it is perfectly possible to change/compensate for the injector size on a diy basis (a carefull done diy basis mind!)
re: the pump load, i just wanted to clarify that simply changing the fuel injector size will not, on its own, reduce the pump load (because the pump is always pumping the same volume of fuel, even if it's not actually being injected into the engine). Reducing the rail setpoint pressure (from 3.5 to 3.0 bar) WILL reduce load as you say, and if you do this, then yes, your correction will need to account for this reduction in flow, i.e.:
The flow increases by ratio of injector capacity (say 218cc/sec to 270cc/sec) then decreases by the SQRT of the pressure reduction (SQRT 3.5bar/3.0bar)
re: the pump load, i just wanted to clarify that simply changing the fuel injector size will not, on its own, reduce the pump load (because the pump is always pumping the same volume of fuel, even if it's not actually being injected into the engine). Reducing the rail setpoint pressure (from 3.5 to 3.0 bar) WILL reduce load as you say, and if you do this, then yes, your correction will need to account for this reduction in flow, i.e.:
The flow increases by ratio of injector capacity (say 218cc/sec to 270cc/sec) then decreases by the SQRT of the pressure reduction (SQRT 3.5bar/3.0bar)
Max_Torque said:
But there is a big difference between an "remap" that generally means tailoring the injection quantity to the engines volumetric efficiency curve, and a "tweak" that is a simple change of injector size. The former, needs a lot of time and some rollers / dyno to do properly, but if you already have the basic mapping done, it is perfectly possible to change/compensate for the injector size on a diy basis (a carefull done diy basis mind!)
re: the pump load, i just wanted to clarify that simply changing the fuel injector size will not, on its own, reduce the pump load (because the pump is always pumping the same volume of fuel, even if it's not actually being injected into the engine). Reducing the rail setpoint pressure (from 3.5 to 3.0 bar) WILL reduce load as you say, and if you do this, then yes, your correction will need to account for this reduction in flow, i.e.:
The flow increases by ratio of injector capacity (say 218cc/sec to 270cc/sec) then decreases by the SQRT of the pressure reduction (SQRT 3.5bar/3.0bar)
Any changes whatsoever to the map, is a remap as far as I am concerned. If you have to delve into the software, you are clearly making adjustments.re: the pump load, i just wanted to clarify that simply changing the fuel injector size will not, on its own, reduce the pump load (because the pump is always pumping the same volume of fuel, even if it's not actually being injected into the engine). Reducing the rail setpoint pressure (from 3.5 to 3.0 bar) WILL reduce load as you say, and if you do this, then yes, your correction will need to account for this reduction in flow, i.e.:
The flow increases by ratio of injector capacity (say 218cc/sec to 270cc/sec) then decreases by the SQRT of the pressure reduction (SQRT 3.5bar/3.0bar)
As I said, doesnt matter if big or small, changes must be made.
And whilst a generic sweep change will make a good starting point, that's all it is. A starting point. The injectors wont behave the same, and adjustments will need to be made at various load sites.
In short, and you know this as well. It does need a remap.
I know you think we need a remap. Maybe we do. But nevertheless, our car is the fastest of it's type in the history of this racing club - only matched by one other car (so far), and that was mostly down to the fact the other driver was very very talented. The engines have never let go. Yes, we have a transient misfire occassionally, but dynos and rolling roads are rubbish at transient mapping anyway. As we run datalogging and a WBO2, we can make educated changes as we go to refine things. This is one such refinement - our injectors are pretty much maxed out prior to any addition multipliers (AE, coolant temp, air temp), so it would make sense to increase their flow rate - and why not reduce pump load at the same time?
Whether I'm right or wrong, stupid or misguided doesn't harm you, so unless you can actually answer my questions then you might as well keep quiet. Repeating "you need a remap" over and over just makes you look sillier, not me.
Thanks for the constructive, helpful, thought-out and patient advice from the others. Very useful and pleased I wasn't a million miles away from being right myself.
Whether I'm right or wrong, stupid or misguided doesn't harm you, so unless you can actually answer my questions then you might as well keep quiet. Repeating "you need a remap" over and over just makes you look sillier, not me.
Thanks for the constructive, helpful, thought-out and patient advice from the others. Very useful and pleased I wasn't a million miles away from being right myself.
tristancliffe said:
I know you think we need a remap. Maybe we do. But nevertheless, our car is the fastest of it's type in the history of this racing club - only matched by one other car (so far), and that was mostly down to the fact the other driver was very very talented. The engines have never let go. Yes, we have a transient misfire occassionally, but dynos and rolling roads are rubbish at transient mapping anyway. As we run datalogging and a WBO2, we can make educated changes as we go to refine things. This is one such refinement - our injectors are pretty much maxed out prior to any addition multipliers (AE, coolant temp, air temp), so it would make sense to increase their flow rate - and why not reduce pump load at the same time?
Whether I'm right or wrong, stupid or misguided doesn't harm you, so unless you can actually answer my questions then you might as well keep quiet. Repeating "you need a remap" over and over just makes you look sillier, not me.
Thanks for the constructive, helpful, thought-out and patient advice from the others. Very useful and pleased I wasn't a million miles away from being right myself.
No, it makes you look stupid.Whether I'm right or wrong, stupid or misguided doesn't harm you, so unless you can actually answer my questions then you might as well keep quiet. Repeating "you need a remap" over and over just makes you look sillier, not me.
Thanks for the constructive, helpful, thought-out and patient advice from the others. Very useful and pleased I wasn't a million miles away from being right myself.
If you think you can fit larger injectors and make NO changes whatsoever to your fuelling maps. Then fair enough. I guess you can, but will the fuelling be correct everywhere ? Not a chance.
Especially since you have already said you are maxing out your current setup. Which means fuelling numbers at the top end are exaggerated due to lack of injector, and you probably have little real control over the fuelling because you current fuel system is inadequate. So switching to a better pump and more suitable injectors could see you far too rich at the top end now thta you actually have proper control over fuelling.
If you cant understand that simple concept, then take the car to someone who does.
Anyone who has ever tuned any cars will know changes MUST be made.
You can call it whatever you want, but what else would you call making adjustments to your map ? re-golf ? re-cricket ?
And changes will be made, as I agree entirely with your reasoning that the fuelling will be out in places. I can alter the maps. When we go testing we can lot AFR and combine that will feel, exhaust smoke, noise (detonation) etc. I will undoubtedly be tweaking the various maps in the ECU to get it back to where it was (which, as you have seen from a previous thread was close, but not quite close enough until I played with it - now we have stable AFRs most of the time).
I don't remember claiming I was going to make no changes whatsoever. In fact, the whole point of this thread is to find out from more experienced people what changes I would have to make. Scholar engines have suggested that running at 100% duty cycle is fine, although I'm inclined to disagree with them based on nobody else I've spoken to echoing it.
But a remap implies taking it to a rolling road or a dyno, and I'm not intending to do that. And if this change doesn't work, it takes about 10 minutes to revert to the current system that we know works.
I don't remember claiming I was going to make no changes whatsoever. In fact, the whole point of this thread is to find out from more experienced people what changes I would have to make. Scholar engines have suggested that running at 100% duty cycle is fine, although I'm inclined to disagree with them based on nobody else I've spoken to echoing it.
But a remap implies taking it to a rolling road or a dyno, and I'm not intending to do that. And if this change doesn't work, it takes about 10 minutes to revert to the current system that we know works.
Edited by tristancliffe on Thursday 6th January 09:44
Does anyone know of compatible injectors between 295cc and 340cc per minute at 3 bar that would be directly replaceable with Bosch 0-280-150-749 EV6 high impedance, top feed 218.1cc/min injectors as used on Rover K-Series engines? The big Stan Weiss list doesn't have much information on what cars used each injector really, so finding a source/supplier is next to impossible.
These would give worst case (i.e. with enrichment multipliers likely to be encountered on very cold days with cooler than normal coolant) duty cycles of 75%-85%, whilst still having sufficient resolution at low RPMs to provide reasonable idle control.
These would give worst case (i.e. with enrichment multipliers likely to be encountered on very cold days with cooler than normal coolant) duty cycles of 75%-85%, whilst still having sufficient resolution at low RPMs to provide reasonable idle control.
Are you looking to buy second hand injectors, or new ?
And I cant think of any new injector that would pose a problem unless the injection system itself is very crude.
Any modern injection system, even running batch fire shouldnt have a problem with resolution, even around 600cc
Sequential injection, you can push that to 1000cc per cylinder with good control.
Certainly a small injector at 350cc will be a doddle.
Assuming the EV6 you are using is the normal "long" length, your options really are massive.
A cheap new injector would be the 42lb ( circa 400cc ) Bosch green. These can be picked up very cheap from the US and will give you plenty of headroom.
US$200 should get you a set, pic taken from this site
http://www.sprayitracing.com/42lb%20Bosch%20EV1%20...
Or Five O has a massive range
http://www.fiveomotorsport.com/domestic-high-perfo...
They list this as an EV6 ?
http://www.fiveomotorsport.com/domestic-high-perfo...
Given brand new are so cheap, I wouldnt even entertain second hand injectors.
And I cant think of any new injector that would pose a problem unless the injection system itself is very crude.
Any modern injection system, even running batch fire shouldnt have a problem with resolution, even around 600cc
Sequential injection, you can push that to 1000cc per cylinder with good control.
Certainly a small injector at 350cc will be a doddle.
Assuming the EV6 you are using is the normal "long" length, your options really are massive.
A cheap new injector would be the 42lb ( circa 400cc ) Bosch green. These can be picked up very cheap from the US and will give you plenty of headroom.
US$200 should get you a set, pic taken from this site
http://www.sprayitracing.com/42lb%20Bosch%20EV1%20...
Or Five O has a massive range
http://www.fiveomotorsport.com/domestic-high-perfo...
They list this as an EV6 ?
http://www.fiveomotorsport.com/domestic-high-perfo...
Given brand new are so cheap, I wouldnt even entertain second hand injectors.
Edited by stevieturbo on Thursday 6th January 11:30
Edited by stevieturbo on Thursday 6th January 11:31
Gassing Station | Engines & Drivetrain | Top of Page | What's New | My Stuff