The more I think about it there is a way to·tell an engine to self calibrate itself and forget about this Mapping process that I just mentioned.· We·already close loop on fuel control at part throttle (O2 sensor).· There are wide range O2 sensors that will read Rich or Lean·A/F ratios,·so all we would have to do is tell EEC what A/F we want to operate at·then through feed-back it could set it.
·The optimum spark advance occurs when the peak cylinder pressure is approx 14° ATDC.· If we closed loop on a·cylinder pressure by means of a pressure sensor·then·spark·timing would always be optimized.·
The problem with both of these concepts is they are expensive; wide range O2 sensors from Horiba around $1000 and cylinder pressure sensors from Kistler in the same range.
GAF, For some reason I thought max HP was developed at 5 degrees ATDC, Is that correct?? Is there a range between optimal and max hp? Couldn't we use a ADXL150? or a knock sensor to determine peak pressure.
Also, have you ever used the 4 stroke model by Gordon Blair, if so is it worth the price tag. I own his 2 stroke model software. In profesional watercraft racing, knowing how to exploit some of his 2 stroke·information, was the difference of·finishing in the top 3 vs the top 10.
GAF and Markistuff,
Their is a company that sells wideband O2 sensors with interface that outputs a Analog voltage from 0-5v and is not to expensive. http://www.innovatemotorsports.com/products/lc1.php they cost $199 and are highly recomended buy alot of the guys I know that tune cars.
I have heard this a number of times, that max power occurs when the peak pressure is at °5 ATDC.·· I believe this idea originally came from·an engine text book by EF Obert, excellent book but he is wrong on that one item.· Over the years I have tested many different types of engines, from fast burn concepts to low compression to high output race engines. Each of these engines were equiped with flush mounted cylinder pressure sensors and a optical encoder, with 0.5 degree resolution, mounted on the front·of the crank.· Virtually all of these produced the same result; that max torque or max power is achieved when spark is adjusted such that, the·location of·peak cylinder pressure is at approx 14° ATDC.· There is a·small range of·peak location over the entire·engine speed range. At 1000 RPM it will be at 14°·and decrease to about·12° at peak power (6000 RPM).· This concept holds true at part throttle also.
Bennettdan,
I·looked at that WEB site and I think that sensor is made by Bosch.· I have seen them used at work, Bosch calls them "Lamdba sensors".· They·might be used in production·somewhere in Europe·which would bring the cost down.· The ones I used were made by NTK and were just for development, thats why they cost·$1000. ··
Yes you can buy the Bosch sensors for around 60 dollars but you have to have the proper circuit to read it, its a little more involved than just a narrow band that you can read with a A/D converter.
It comes with the bosch O2 sensor and everything you need.
But is you want just the wide band O2 sensor alone they can be bought at most autoparts stores the Bosch part number is 0 258 007 057 or 0 258 007 058·for a sensor that was uses on some Voltswagons and they run around 30 to 60 dollars.
Post Edited (bennettdan) : 6/12/2007 2:31:54 AM GMT
Okay i know most of yall r a bit sensitive to critizism, but i'll try to be constructive... and i would like to formerly welcome you all to my thread i'll be starting on my version of a DFI in the Propeller forum.
Awesome effort here guys!! Good points made in this thread and there were some helpful formula's over on the BS2 linked thread. Both SX and BS2 ECU's would benefit greatly from a "wide-band" O2 sensor however, allowing precision readings that are alot more linear than the "cheapy" O2 versions. Think i'm full of BS? Look up the datasheets for a standard "cheapy" O2 sensor and look at the linearity curve for voltage vs temp, it's a joke for reading stoich smoothly! The curve is so steep it's more like a high-speed comparator than an accurate thermocoupling device, yeah T/C, it doesn't sense Oxygen levels at all in case you were wondering. It's just a calibrated "high speed" thermocouple, and the cheap ones are a poor one at that.
A Wide-band version may cost more but the temp/volt curve is ALOT more linear giving you alot more presision that leads to a smoother running, better responding engine. I have a Buick Grand National on the stock ECU/block reading a wideband O2 provided through a bit of electronic conditioning and it runs very smooth, even if you "smack" it to the floor, currently the car is in the low 11's runing an 11.1 @ 121MPH w/a 60ft time of 1.5 sec, that's ~avg 550HP, i say average cause i was on/off the gas all the way down the 1/4, it's never been dyno'd, only calculated. I know now after looking at the numbers that the turbine housing was a bit small "choking" the engine causing it to break-up and mis-fire.
75MHz is not overkill for EFI, in fact, it's what most aftermarket DFI's are running close to, some are running @ 120MHz core speed "xFAST FI" or DFI i think it's called, it was at SEMA last yr - (Digital Sequential Fuel Injection).
Please view my thread in the propeller forum on this project, or subscribe to it if you like. But i'm almost expecting some critisism on my efforts, I'm hoping all your suggestions will inevitably lead to improvements as a set of fresh eyes will always have a different point of view.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Definetly a E3 (Electronics Engineer Extrodinare!)
"I laugh in the face of imposible,... not because i know it all, ... but because I don't know well enough!"
bennettdan - my older bro ordered a wide band o2 sensor, and he figured i could build a controller for him with the sx that would tell him what air fuel ratio he is running. he wants it for tunning the carb on his cobra. i orignaly thought the wide band worked of of off of 0 - 5 volt range for telling you the AFR but apparently it uses current one way or another to equate to the amount of air in the exsaust. you wouldnt have any idea how to go about building a controller board would you? or should i just tell him to buy one. yes, ill use it as well for tunnin my fuel injection. trial and error is a pain in the butt and takes along time.
Do you have a datasheet on the Wide band O2 sensor? If it is a regular bosch check out this sight I have not even started on a wideband controller. I have the LC-1 from innovative for now it outputs a 0-5v analog signal to be read by a ECU or you can make a handheld. http://wbo2.com/lsu/lsuworks.htm
One more thing as far as I can tell the only difference between 5 wire and 6 wire sensors is that the 6 wire sensors just have a resistor calibration pin for freeair calibration, but I am not totally sure about that.
Post Edited (bennettdan) : 6/15/2007 3:16:27 AM GMT
K&N air filters sells a montor for jetting purposes
GAF,
About ignition timing again, When I think of Ford, I think of V8's. But I believe your statement included inline 4's like the older 2000 & 2300's that were in Pinto's and Mustang 2's? Also I was wondering about that relationship to the Rod-Length ratio.·Do most car engines run Rod ratios above 1.95 : 1?, is it possible that old text was referring to an engine with a rod ratio of 1.85:1 or less?
Many of my model airplane, scale·pylon racing engines were converted chain·saw engines, many·were large cc singles converted to 200cc twin cylinder opposed or older twins, up to 290cc that had been converted to UAV drone engines (Like a very large Kenmore washer engine [noparse]:)[/noparse]. Nearly all these engines have a 1.65:1 rod ratio, Does it make sense to you, that it would be better to have peak pressure closer to the 5 deg atdc on these engines?
A complete engine management system is my goal for these race engines, but I don't know if I will use Spark Ignition or Glow Plugs yet. All attempts with SI so far has failed to beat Glow Plugs, but there are too many variables that effect glow plug timing vs SI. My previous experience with inline twin gasoline SI watercraft engines and rod ratios of aprox. 1.95:1, was 28 degrees btdc worked best. But now I am thinking I should try closer to 40 degrees BTDC on my 1.65:1 model airplane engines.
Bill
Post Edited (Capt. Quirk) : 6/15/2007 4:59:35 PM GMT
You are correct that Rod ratios for automotive applications·are approx 1.8:1.· The length of the rod is dictated by many variables and trade-offs.· A rod that is·short takes advantage of higher gas pressure earlier on the expansion stroke, the trade off here is the increased side loading on the piston, which increases engine friction, so there is no real gain.· I suppose the best way to get a handle on this is to look at the equation that relates cylinder volume to crank position (this includes the rod length). I have set up a little program to·perform this calculation.· From this data you can plot cylinder volume vs crank angle with various rod lengths.· To see how much effect the rod·has on this, I ran two cases: (1.)·6" Rod· and (2.) 12" Rod.· When I plot thes two cases the lines are so close together you can hardly see the difference.· So for automotive engines a fairly long rod gives the advantage of lower friction and better fuel ecomomy.·
For peak pressure location,·I have looked at·many different engines over my career as a engineer·and they all (independent of size and geometry) are optimized when the peak cylinder pressure is at ~14° ATDC.· I would suspect, with a glow plug system,·the·mixture·probably fires around TDC, yielding a peak pressure that is not· optimum, (it would be interesting to measure this).· It would be a good project to at least get the ignition system working for one of these engines that·uses a glow plug, just to see if a single spark will beat a glow plug system.·Try to get a long spark duration (3 ms).
i havent really made to much progress lately.· its been running real good for about a month now. no real problems with the ecu.· i havent hooked up the electronic igintion yet·or finished fine tunning the program code,·but it runs way better then the carb i had on it.· so good infact ive broken 2 chains and·im haveing real problems with my transmission.· guess with the engine is makeing more power now with efi so the other parts of the cart cant keep up.· thats mainly why i havent done more work on the ecu.· and im in the middle of building a wide band o2 controller.· almost have it done.· i think its alittle more complicated for me to build then the ecu was.
Comments
The more I think about it there is a way to·tell an engine to self calibrate itself and forget about this Mapping process that I just mentioned.· We·already close loop on fuel control at part throttle (O2 sensor).· There are wide range O2 sensors that will read Rich or Lean·A/F ratios,·so all we would have to do is tell EEC what A/F we want to operate at·then through feed-back it could set it.
·The optimum spark advance occurs when the peak cylinder pressure is approx 14° ATDC.· If we closed loop on a·cylinder pressure by means of a pressure sensor·then·spark·timing would always be optimized.·
The problem with both of these concepts is they are expensive; wide range O2 sensors from Horiba around $1000 and cylinder pressure sensors from Kistler in the same range.
Also, have you ever used the 4 stroke model by Gordon Blair, if so is it worth the price tag. I own his 2 stroke model software. In profesional watercraft racing, knowing how to exploit some of his 2 stroke·information, was the difference of·finishing in the top 3 vs the top 10.
Their is a company that sells wideband O2 sensors with interface that outputs a Analog voltage from 0-5v and is not to expensive.
http://www.innovatemotorsports.com/products/lc1.php they cost $199 and are highly recomended buy alot of the guys I know that tune cars.
I have heard this a number of times, that max power occurs when the peak pressure is at °5 ATDC.·· I believe this idea originally came from·an engine text book by EF Obert, excellent book but he is wrong on that one item.· Over the years I have tested many different types of engines, from fast burn concepts to low compression to high output race engines. Each of these engines were equiped with flush mounted cylinder pressure sensors and a optical encoder, with 0.5 degree resolution, mounted on the front·of the crank.· Virtually all of these produced the same result; that max torque or max power is achieved when spark is adjusted such that, the·location of·peak cylinder pressure is at approx 14° ATDC.· There is a·small range of·peak location over the entire·engine speed range. At 1000 RPM it will be at 14°·and decrease to about·12° at peak power (6000 RPM).· This concept holds true at part throttle also.
Bennettdan,
I·looked at that WEB site and I think that sensor is made by Bosch.· I have seen them used at work, Bosch calls them "Lamdba sensors".· They·might be used in production·somewhere in Europe·which would bring the cost down.· The ones I used were made by NTK and were just for development, thats why they cost·$1000.
··
Do you know where I could buy the Bosch Wide Range Sensor and the driver board?
You can buy the LC-1 from this sight
http://www.innovatemotorsports.com/products/lc1.php
It comes with the bosch O2 sensor and everything you need.
But is you want just the wide band O2 sensor alone they can be bought at most autoparts stores the Bosch part number is 0 258 007 057 or 0 258 007 058·for a sensor that was uses on some Voltswagons and they run around 30 to 60 dollars.
Post Edited (bennettdan) : 6/12/2007 2:31:54 AM GMT
Awesome effort here guys!! Good points made in this thread and there were some helpful formula's over on the BS2 linked thread. Both SX and BS2 ECU's would benefit greatly from a "wide-band" O2 sensor however, allowing precision readings that are alot more linear than the "cheapy" O2 versions. Think i'm full of BS? Look up the datasheets for a standard "cheapy" O2 sensor and look at the linearity curve for voltage vs temp, it's a joke for reading stoich smoothly! The curve is so steep it's more like a high-speed comparator than an accurate thermocoupling device, yeah T/C, it doesn't sense Oxygen levels at all in case you were wondering. It's just a calibrated "high speed" thermocouple, and the cheap ones are a poor one at that.
A Wide-band version may cost more but the temp/volt curve is ALOT more linear giving you alot more presision that leads to a smoother running, better responding engine. I have a Buick Grand National on the stock ECU/block reading a wideband O2 provided through a bit of electronic conditioning and it runs very smooth, even if you "smack" it to the floor, currently the car is in the low 11's runing an 11.1 @ 121MPH w/a 60ft time of 1.5 sec, that's ~avg 550HP, i say average cause i was on/off the gas all the way down the 1/4, it's never been dyno'd, only calculated. I know now after looking at the numbers that the turbine housing was a bit small "choking" the engine causing it to break-up and mis-fire.
75MHz is not overkill for EFI, in fact, it's what most aftermarket DFI's are running close to, some are running @ 120MHz core speed "xFAST FI" or DFI i think it's called, it was at SEMA last yr - (Digital Sequential Fuel Injection).
Please view my thread in the propeller forum on this project, or subscribe to it if you like. But i'm almost expecting some critisism on my efforts, I'm hoping all your suggestions will inevitably lead to improvements as a set of fresh eyes will always have a different point of view.
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
Definetly a E3 (Electronics Engineer Extrodinare!)
"I laugh in the face of imposible,... not because i know it all, ... but because I don't know well enough!"
I don't see you thread ?
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
*.*
Ibsen
" It's nice to be important, but
·· more important to be nice... "
http://wbo2.com/lsu/lsuworks.htm
One more thing as far as I can tell the only difference between 5 wire and 6 wire sensors is that the 6 wire sensors just have a resistor calibration pin for freeair calibration, but I am not totally sure about that.
Post Edited (bennettdan) : 6/15/2007 3:16:27 AM GMT
K&N air filters sells a montor for jetting purposes
GAF,
About ignition timing again, When I think of Ford, I think of V8's. But I believe your statement included inline 4's like the older 2000 & 2300's that were in Pinto's and Mustang 2's? Also I was wondering about that relationship to the Rod-Length ratio.·Do most car engines run Rod ratios above 1.95 : 1?, is it possible that old text was referring to an engine with a rod ratio of 1.85:1 or less?
Many of my model airplane, scale·pylon racing engines were converted chain·saw engines, many·were large cc singles converted to 200cc twin cylinder opposed or older twins, up to 290cc that had been converted to UAV drone engines (Like a very large Kenmore washer engine [noparse]:)[/noparse]. Nearly all these engines have a 1.65:1 rod ratio, Does it make sense to you, that it would be better to have peak pressure closer to the 5 deg atdc on these engines?
A complete engine management system is my goal for these race engines, but I don't know if I will use Spark Ignition or Glow Plugs yet. All attempts with SI so far has failed to beat Glow Plugs, but there are too many variables that effect glow plug timing vs SI. My previous experience with inline twin gasoline SI watercraft engines and rod ratios of aprox. 1.95:1, was 28 degrees btdc worked best. But now I am thinking I should try closer to 40 degrees BTDC on my 1.65:1 model airplane engines.
Bill
Post Edited (Capt. Quirk) : 6/15/2007 4:59:35 PM GMT
You are correct that Rod ratios for automotive applications·are approx 1.8:1.· The length of the rod is dictated by many variables and trade-offs.· A rod that is·short takes advantage of higher gas pressure earlier on the expansion stroke, the trade off here is the increased side loading on the piston, which increases engine friction, so there is no real gain.· I suppose the best way to get a handle on this is to look at the equation that relates cylinder volume to crank position (this includes the rod length). I have set up a little program to·perform this calculation.· From this data you can plot cylinder volume vs crank angle with various rod lengths.· To see how much effect the rod·has on this, I ran two cases: (1.)·6" Rod· and (2.) 12" Rod.· When I plot thes two cases the lines are so close together you can hardly see the difference.· So for automotive engines a fairly long rod gives the advantage of lower friction and better fuel ecomomy.·
For peak pressure location,·I have looked at·many different engines over my career as a engineer·and they all (independent of size and geometry) are optimized when the peak cylinder pressure is at ~14° ATDC.· I would suspect, with a glow plug system,·the·mixture·probably fires around TDC, yielding a peak pressure that is not· optimum, (it would be interesting to measure this).· It would be a good project to at least get the ignition system working for one of these engines that·uses a glow plug, just to see if a single spark will beat a glow plug system.·Try to get a long spark duration (3 ms).
▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔▔
*.*
Ibsen
" It's nice to be important, but
·· more important to be nice... "