As I mentioned in another thread (Auxiliary Power sockets) I now have a reliable ignition on 12v power supply and got around to connecting up the Tech Edge wide band oxygen sensor system so I know whats going on mixture wise . As mentioned previously when the new engine went in I had my fabricator fit another O2 probe fitting and plug so I can run the wide band probe without removing the std computers one .
We went for a burn south on the Princes Hwy to see what sort of mixture ratios Ellie has and in many ways its better than I thought it would be .
Hot it idles between about 14.8 and 15 to 1 AFR which was about right for pre historic emissions cars . For most cruising around town it sits in the 13-13.5 to 1 and at highway speeds of 100-110 about the 14.7 to 14.85 to 1 AFR .
Uphill in high gears and WOT (wide open throttle) it goes pig rich like 11 or 10 to 1 and if you hold it flat it goes off the scale at under 10 to one which is mega rich .
On the down side it should run richer in the upper medium throttle/load ranges but wafts along still in the 14.5/14.8 range . Its possible that the TPS open throttle contacts are a bit dicky or a turbo/airflow problem because my car never was the same boost wise since the turbo came off to fix it's leaking lower water pipe .
Anyhow when it gets a better turbo header and a programmable computer this thing should get along a fair bit better than it does now .
At least we can now see what it's air fuel ratios are doing , cheers A .
Finally got the Tech Edge Wide Band working .
-
discopotato03
- Senior Member
- Posts: 2134
- Joined: Sun Mar 18, 2007 9:29 am
- Location: Sydney
-
steptoe
- Master Member
- Posts: 11582
- Joined: Thu Oct 06, 2005 10:00 am
- Location: 14 miles outside Gotham City
Gee's. Interesting to see it run that rich under boost and that is where O2 sensor is not involved in the air fuel ratio. Will your new computer be controlling fuel ratio according to O2 sensor even under boost? As a comparison with same engime in mine but LPG where stoich is 15.5: instead of 14.7:1 for petrol Ive had boost mix of 12.5 - 13:1 but now run 13.2 - 13.8:1 for turbo preservation given that the richer the mixer the hotter things become (like Diesels) whereas the richer in petrol things are a little cooler.
-
discopotato03
- Senior Member
- Posts: 2134
- Joined: Sun Mar 18, 2007 9:29 am
- Location: Sydney
I'm not 100% certain but it may have something to do with the turbo not boosting to full (std) pressure so with less air at wide open throttle under load it's possible for the mixture to go very rich .
I certainly wouldn't tune an engine any richer than 11 to 1 on petrol because with reasonably modern computers you get much more accurate and there fore tighter control of engine fueling and ignition timing .
By this I mean to use the later optical CAS and sequential fuel injection .
Computers have developed to the stage where MAP sensing can be good enough though mass air sensing is nice if possible .
I believe what the Vipec and Link G4 computers do is throttle position sensing fuel control with MAP correction to tidy it all up .
The 3 plug Subaru system is more crude than I initially thought and won't do what I need it too . Where it's more basic than the FJ20's management is the 2 stage batch fire injection .
Most people don't try to run in closed loop except for light loads , at full load you can usually have reasonably accurate control of AFR (provided the pumps and injectors are up to the task) and its preferable to let it run a tad rich than a tad lean . You have to be a bit careful because slight differences in manifolding and head porting can make individual cylinders run leaner (hotter) or richer (cooler) so you have to cater for the leanest/hottest cylinder because invariably it's the one that wants to detonate first .
Engines with individual exhaust ports and manifold runners make it easy to run an exhaust gas temp sensor in each one and decent computers have correction features to trim mixtures in individual cylinders . Then if necessary you can run closer to 12 and leaner AFR's and get the nth degree of power from a given engine . This is really only important with competition cars that get limited fuel stops so fuel consumption is critical if the car is to finish the event .
For roadies where reliability is king having a tad worse fuel consumption at full load is a small price to pay because you don't spend long periods "on the floor" so to speak .
I tend to think that mixtures between 13.5 - 14.7 for idle and light load running and down to maybe 11.0 should work well for me . Being able to reference ignition timing off a map sensor rather than bob weights and a vacuum can should be a huge advantage giving the thing much better throttle response and better torque from more advance and similar AFR's .
I'd like it to be able to get better than ten to the hundred fuel consumption as well .
Cheers A .
I certainly wouldn't tune an engine any richer than 11 to 1 on petrol because with reasonably modern computers you get much more accurate and there fore tighter control of engine fueling and ignition timing .
By this I mean to use the later optical CAS and sequential fuel injection .
Computers have developed to the stage where MAP sensing can be good enough though mass air sensing is nice if possible .
I believe what the Vipec and Link G4 computers do is throttle position sensing fuel control with MAP correction to tidy it all up .
The 3 plug Subaru system is more crude than I initially thought and won't do what I need it too . Where it's more basic than the FJ20's management is the 2 stage batch fire injection .
Most people don't try to run in closed loop except for light loads , at full load you can usually have reasonably accurate control of AFR (provided the pumps and injectors are up to the task) and its preferable to let it run a tad rich than a tad lean . You have to be a bit careful because slight differences in manifolding and head porting can make individual cylinders run leaner (hotter) or richer (cooler) so you have to cater for the leanest/hottest cylinder because invariably it's the one that wants to detonate first .
Engines with individual exhaust ports and manifold runners make it easy to run an exhaust gas temp sensor in each one and decent computers have correction features to trim mixtures in individual cylinders . Then if necessary you can run closer to 12 and leaner AFR's and get the nth degree of power from a given engine . This is really only important with competition cars that get limited fuel stops so fuel consumption is critical if the car is to finish the event .
For roadies where reliability is king having a tad worse fuel consumption at full load is a small price to pay because you don't spend long periods "on the floor" so to speak .
I tend to think that mixtures between 13.5 - 14.7 for idle and light load running and down to maybe 11.0 should work well for me . Being able to reference ignition timing off a map sensor rather than bob weights and a vacuum can should be a huge advantage giving the thing much better throttle response and better torque from more advance and similar AFR's .
I'd like it to be able to get better than ten to the hundred fuel consumption as well .
Cheers A .
-
steptoe
- Master Member
- Posts: 11582
- Joined: Thu Oct 06, 2005 10:00 am
- Location: 14 miles outside Gotham City
I always thought that to be more to the lean produces best power with petrol.
From memory, the Kluger has an engine capacity twice that of our EA82T's, twice the power, vehicle is twice the weight and gets the same economy or thereabouts .
I'm sure all the commercial vehicles over last 20 years have done same double capacity and power and retained same fuel economy, something impossible for us to do with these old dinosaurs.
From memory, the Kluger has an engine capacity twice that of our EA82T's, twice the power, vehicle is twice the weight and gets the same economy or thereabouts .
I'm sure all the commercial vehicles over last 20 years have done same double capacity and power and retained same fuel economy, something impossible for us to do with these old dinosaurs.
-
discopotato03
- Senior Member
- Posts: 2134
- Joined: Sun Mar 18, 2007 9:29 am
- Location: Sydney
I'd like to try the real world experience with a Kluger , as in my weekly travels to see if it could get around 10/100 . I tend to think with the traffic driving I do it wouldn't get within a bulls roar of 10/100 . Don't think that I do much freeway running in my Subie because I don't .
Well the chemically correct mixture by weight or mass is ~ 14.7 to 1 air to fuel . A lot of competition development has gone into fuel consumption because of the number of fuel stops allowed in some events . I remember being told about some exotic turbo Nissan in a 24 hour event that never ran richer than thirteen point something to one - for the whole event .
There are degrees of "lean" and it is really a perception thing . 14.7 to 1 is not really lean because in theory burning at that ratio uses all the fuel and all the oxygen . Technically anything richer gives an excess fuel factor and vise versa with oxygen running leaner ratios . Later developments have moved towards an excess air factor or leaner than 14.7 to 1 to push consumption and to varying degrees "clean emissions" . Doing things like exhaust gas recirculation is aimed at keeping compustion temps down when running in excess air conditions - it reduces NOx by keeping temps below the point where nitrogen wants to bond with oxygen and also helps prevent detonation .
Aside from reasons of cost production engines could have nock and exhaust temp sensor for each cylinder and could probably run leaner average mixtures if required . I think they are concentrating on direct chamber injection (GDI) at the moment and this enables them to get some diesel engine like characteristics one of which is injecting just before TDC on the compression stroke and getting low specific fuel consumption .
I have no doubt whatsoever that better than original fuel consumption can be had from an EA82T specifically the vane AFM ones like mine because its engine management is so crude in std form . Pump fuel has come a long way from the lets go ULP rubish we got back in 1985 but the crude systems don't really let you take full advantage of it . As I mentioned a mechanical distributor ignition system cannot emulate what an electronically timed system can do to achieve best possible combustion characteristics .
Also the twin switch TPS system can't tell the computer exactly how far open the throttle actually is , all it signals is closed/light load/high load .
People want to argue the toss over sequential vs batch fire injection but virtually every production car and half serious competition engine has it so there IS something in it . The IS I believe is having control over injection timimg in the 4 stroke or 720 degree cycle . It injects sequentially on a cylinder by cylinder basis so the AFR transients have to be better . With batch fire the injection timing is all over the place with complete disregard to inlet valve timing and the result is that mixture transients aren't as rapid as engine load and speed transients . Average mixtures tend to be a littlte richer for no other reason that less accurate controls .
One thing you do find currently is that manufacturers set their systems to run rich and retarded under high load situations and I don't have an answer to that one . I suppose it's a reliability with adequate emissions thing but not the best method of getting best power per pound of fuel .
Anyhow one of my challenges will be to get over 30 mpg but I'm not going to run it lean and hot to get there because I'm not risking reliability - with EA82 heads ...
Cheers A .
Well the chemically correct mixture by weight or mass is ~ 14.7 to 1 air to fuel . A lot of competition development has gone into fuel consumption because of the number of fuel stops allowed in some events . I remember being told about some exotic turbo Nissan in a 24 hour event that never ran richer than thirteen point something to one - for the whole event .
There are degrees of "lean" and it is really a perception thing . 14.7 to 1 is not really lean because in theory burning at that ratio uses all the fuel and all the oxygen . Technically anything richer gives an excess fuel factor and vise versa with oxygen running leaner ratios . Later developments have moved towards an excess air factor or leaner than 14.7 to 1 to push consumption and to varying degrees "clean emissions" . Doing things like exhaust gas recirculation is aimed at keeping compustion temps down when running in excess air conditions - it reduces NOx by keeping temps below the point where nitrogen wants to bond with oxygen and also helps prevent detonation .
Aside from reasons of cost production engines could have nock and exhaust temp sensor for each cylinder and could probably run leaner average mixtures if required . I think they are concentrating on direct chamber injection (GDI) at the moment and this enables them to get some diesel engine like characteristics one of which is injecting just before TDC on the compression stroke and getting low specific fuel consumption .
I have no doubt whatsoever that better than original fuel consumption can be had from an EA82T specifically the vane AFM ones like mine because its engine management is so crude in std form . Pump fuel has come a long way from the lets go ULP rubish we got back in 1985 but the crude systems don't really let you take full advantage of it . As I mentioned a mechanical distributor ignition system cannot emulate what an electronically timed system can do to achieve best possible combustion characteristics .
Also the twin switch TPS system can't tell the computer exactly how far open the throttle actually is , all it signals is closed/light load/high load .
People want to argue the toss over sequential vs batch fire injection but virtually every production car and half serious competition engine has it so there IS something in it . The IS I believe is having control over injection timimg in the 4 stroke or 720 degree cycle . It injects sequentially on a cylinder by cylinder basis so the AFR transients have to be better . With batch fire the injection timing is all over the place with complete disregard to inlet valve timing and the result is that mixture transients aren't as rapid as engine load and speed transients . Average mixtures tend to be a littlte richer for no other reason that less accurate controls .
One thing you do find currently is that manufacturers set their systems to run rich and retarded under high load situations and I don't have an answer to that one . I suppose it's a reliability with adequate emissions thing but not the best method of getting best power per pound of fuel .
Anyhow one of my challenges will be to get over 30 mpg but I'm not going to run it lean and hot to get there because I'm not risking reliability - with EA82 heads ...
Cheers A .