60lb Inj. Problem

[SJSchafer]

The actual calculations are done on a set of 4 dynamnically flow matched siemens 60s on a flow bench with special algorithms written to find the dead time calculations (offsets), linear PW ranges, the adder values needed to bring the injectors back to their linear range in the low PW areas (which for these injectors are all values under 3.75ms) and other info. They are tested using actual gasoline at 120*f. The tests were done for me at 58 psi from 10-16 volts at -120 - 80 kPa VAC. Since I have a returnless system at a constant pressure of 58 psi, I just need to figure out the difference between 58psi@100kPa and 48psi@54kPa for the net difference in kPa between our fuel pressures and I can copy the proper "row" from my table. Just math.

[SJSchafer]

Okay, BlownZ06, here are your offsets:

Volts offset
4.5- 8.55103
5- 7.56836
5.5- 6.57959
6- 5.60303
6.5- 4.63257
7- 3.67432
7.5- 2.71606
8- 1.77002
8.5- 1.5564
9- 1.36108
9.5- 1.18408
10- 1.03149
10.5- 0.90332
11- 0.79346
11.5- 0.7019
12- 0.62256
12.5- 0.55542
13- 0.50049
13.5- 0.45776
14- 0.41504
14.5- 0.36621
15- 0.32349
15.5- 0.28687
16- 0.25024
16.5- 0.26245
17- 0.26855
17.5- 0.26855
18- 0.26245
18.5- 0.24414
19- 0.21362
19.5- 0.177
20- 0.12817
20.5- 0.07935

[SJSchafer]

Code4, here are yours:

Volts Offset
4.5- 8.55103
5- 7.55615
5.5- 6.57349
6- 5.60303
6.5- 4.64478
7- 3.69873
7.5- 2.76489
8- 1.83716
8.5- 1.61133
9- 1.40991
9.5- 1.22681
10- 1.06812
10.5- 0.93994
11- 0.83618
11.5- 0.74463
12- 0.66528
12.5- 0.58594
13- 0.5188
13.5- 0.45776
14- 0.40894
14.5- 0.37231
15- 0.33569
15.5- 0.31738
16- 0.30518
16.5- 0.30518
17- 0.29907
17.5- 0.29297
18- 0.28076
18.5- 0.26245
19- 0.23193
19.5- 0.18921
20- 0.14038
20.5- 0.08545

[Blackbird]

SJ your PM box is full.

[BLOWNZO6]

Okay, BlownZ06, here are your offsets:

Volts offset
4.5- 8.55103
5- 7.56836
5.5- 6.57959
6- 5.60303
6.5- 4.63257
7- 3.67432
7.5- 2.71606
8- 1.77002
8.5- 1.5564
9- 1.36108
9.5- 1.18408
10- 1.03149
10.5- 0.90332
11- 0.79346
11.5- 0.7019
12- 0.62256
12.5- 0.55542
13- 0.50049
13.5- 0.45776
14- 0.41504
14.5- 0.36621
15- 0.32349
15.5- 0.28687
16- 0.25024
16.5- 0.26245
17- 0.26855
17.5- 0.26855
18- 0.26245
18.5- 0.24414
19- 0.21362
19.5- 0.177
20- 0.12817
20.5- 0.07935

Thanks if you lay them in the table they are exactly the same from 13-15 volts as what I already had in there....

[SJSchafer]

SJ your PM box is full.

Emptied.

[SJSchafer]

Thanks if you lay them in the table they are exactly the same from 13-15 volts as what I already had in there....

Sweet.

[BLOWNZO6]

Sweet.

I am going to swap out a pump tomorrow and run up the base fuel pressure a bit so I think these will work still.

Just comparing them to what you put in the other guys table.

[4.8t]

wow this is alot to read. i may run 60's also. 04 tahoe.

[John_D.]

Great info. I've been searching for the offset data for the injectors for months. Just found this info yesterday. Plugged the offset numbers in last night. They were actually pretty close to what I had in the typical operating voltage range, but I still feel better having something that's based on actual measured injector performance.

When you get some good SPA numbers, I'd love to get my hands on those too!

Previously I had just filled in the SPA table so that anything under 1.1 msec would be brought up to 1.1 msec. Good to know that the SPA table is actually used to make the injectors appear linear to the PW calc, and to see some definite info on where the linearity starts to fall off.

[4.8t]

here is my starter tune for the 60's. what else besides offset should i change?

[joffa]

Any chance i could get these offsets for 65psi rail pressure? my car idles at 58kpa map. Thanks in advance.

[4.8t]

Sweet.

did you ever get any good adder info? i was trying to tune my ve/maf some tonight and when i disable trims and all that, nothing i do makes a difference. i even tried my RTT and raised and lowered numbers and it wouldn't do anything.

[4.8t]

so let me see if i understand this now. short pulse limit tells the pcm " hey when you get below this add the corrisponding short pulse adder" so say your short pulse limit it 3.00 and you hit a 1.5 then it takes 1.5 and adds the offset value and the short pulse adder value. the minimum injector pulse is the lowest the pcm will let itself tell the injector to fire. but a value of say .745 is not really a pulse of .745 but rather a pulse of .745 plus the offset value plus the short pulse adder being that .745 is lower than the short pulse limit of 3.00. if this is true it would explain why i can't get a leaner idle. because time you add all this up with my min inj pulse at .745 you really get 1.8ms which is rich. but if i lower the min inj pulse and get the adders where they will stablize the injector at such low pulse then i can go leaner without missfire. because the adder would be like "hey dummy, you can't fire that fast so add this to your time". does this make since? or am i way out in left field here.


this would also explain why when i try to tune the ve or maf for idle with RTT none of the changes to make it leaner would do anything. because after it was all add up 1.8ms was as low as it could possibly go.

[SJSchafer]

so let me see if i understand this now. short pulse limit tells the pcm " hey when you get below this add the corrisponding short pulse adder" so say your short pulse limit it 3.00 and you hit a 1.5 then it takes 1.5 and adds the offset value and the short pulse adder value. the minimum injector pulse is the lowest the pcm will let itself tell the injector to fire. but a value of say .745 is not really a pulse of .745 but rather a pulse of .745 plus the offset value plus the short pulse adder being that .745 is lower than the short pulse limit of 3.00. if this is true it would explain why i can't get a leaner idle. because time you add all this up with my min inj pulse at .745 you really get 1.8ms which is rich. but if i lower the min inj pulse and get the adders where they will stablize the injector at such low pulse then i can go leaner without missfire. because the adder would be like "hey dummy, you can't fire that fast so add this to your time". does this make since? or am i way out in left field here.


this would also explain why when i try to tune the ve or maf for idle with RTT none of the changes to make it leaner would do anything. because after it was all add up 1.8ms was as low as it could possibly go.

That is the basic principal. From the adder tuning I have so far, I have a sloped curve (sort of like the MAF curve) from .0391 at 1ms to .6172 at 0ms. Then I found my "pintle bounce" area to be at about the .750 area, so my slpoe actually goes from 1ms to .750, then drops at .625 and slopes to 0ms. Sort of like this:

Labels 0.000 0.125 0.250 0.375 0.500 0.625 0.750 0.875 1.000
Pulse Width Adder 0.61719 0.50781 0.39844 0.29688 0.19531 0.11719 0.14844 0.08594 0.03906

I still have a little roughness at idle, which I hope to smooth out in the next few days. One thing that I have found is that the pintle bounce changes from injector to injector. My last set was at .625.

Set the nim pw really low, so you don't run into a wall...

[redhardsupra]

hey steve, wanna post your methodology on SPA tuning?

[4.8t]

so what is pintle bounce? maybe i missed something but how do you get pintle bounce on disc injectors? not being a smartass just wondering.

well i'm realy looking forward to what you come up with on adders. i'm still playing with mine to see what i can get but it would be nice to see what you get.

[SJSchafer]

hey steve, wanna post your methodology on SPA tuning?

Hey Marcin. Sure. What I am doing is I am starting with a known IFR and offset. Then I set the plot filter in my MAF histo to only log 1.5ms and up. Do to the change in offsets and IFR between accel and decel you will get cells that command different PWs in the same cell. For instance, in the 2100hz cell, I command around.997ms at idle. In hard decel I command as low as .671 in the same 2100hz cell.

So getting back to it, you set your histo filter to 1.5 and up and tune your MAF (as much as you can). Then you create a histo that logs your PWs to 3 decimal places. Check your low PWs when in the MAF cells that are already dialed in. If my MAF is dailed at 3600 hz at 1.5ms during accel, but runs way lean during decel at .900ms, then I have a fueling error. I do the same thing working my way down untill I am at a point where I am just working on the really low areas around .6-.7ms, decreasing my histo filter to 1.2ms and eventually to .9ms.

One of the ways I set things up is that I create 2 VE table histos. One logging AFR error and one PWs. Even though I am in MAF mode and cannot change the VE based on the info I have, I can still see where my rich and lean errors are and what PWs I am running when said errors occur. I also get MAF dialed in and then switch to VE mode to double check my work. MAF is quite a bit more forgiving. On more than one occasion I have thought I had it really close and switched to VE only to show that I was off a bit. The trick there is to notice when you go from 16.xx to 12.xx in the same VE cell. Then you have to ask yourself when adjusting if, by looking at the other cells, does that huge peak/valley belong there? Keep in mind that most errors are going to be between adder cells. When you go from lean to rich and back to lean, it can mean that one cell is too high or one too low or both.

The biggest problem I have is that, due to being on a returnless system, I have to force these injectors down to a point where they just don't want to flow at all. My next mod when I get some money together is a return system.

[redhardsupra]

good stuff, but i got some questions (you didn't think i'd let you off easy did ya? )
1. how do you know when to attribute changes to MAF and when to SPA?
2. how do you make MAF agree with VE?
3. why use 1.5ms as the filter for data gathering? do you know that 60s flow linearly past that point?
4. how do you know that your IFR and offset are 'known'? do you measure dynamic fuel pressure? did you get reliable offset data from Paul Yaw? I thought there was still some fuzzy areas with these...
5. why view the same data on AFR and IPW scales? since the two are connected, why not use them both to backcalculate airmass and compare it with the figure obtained from MAF?

[SJSchafer]

good stuff, but i got some questions (you didn't think i'd let you off easy did ya? )
1. how do you know when to attribute changes to MAF and when to SPA?
2. how do you make MAF agree with VE?
3. why use 1.5ms as the filter for data gathering? do you know that 60s flow linearly past that point?
4. how do you know that your IFR and offset are 'known'? do you measure dynamic fuel pressure? did you get reliable offset data from Paul Yaw? I thought there was still some fuzzy areas with these...
5. why view the same data on AFR and IPW scales? since the two are connected, why not use them both to backcalculate airmass and compare it with the figure obtained from MAF?

The answer to 1 and 3 are basically the same. The reason that you filter the MAF at 1.5 is precisely because the injectors flow (mostly) linearly there. That gives you the opportunity to dial in your MAF excluding injector errors. Once the MAF is tuned with that method, any errors in AFR in low PW areas in those areas of the MAF that was dialed in through the method above can only be fueling errors. You will only have the cells from, 3300hz (on my car) and up to work with, but that will allow you to dial in your adders to the point where you can lower your filter to 1.1ms to dial in the 2400 and 2800. At that point, go to .9 on your filter and dial in your 2100 and 1800. Make sense?

4 is easy. Yes, I have had the injectors dynamically flow tested for IFRs and offsets, using gasoline at 120 degrees f. The only gray area is the adders as the change from injector to injector.

MAF and VE don't have to agree. You are just looking at the VE AFR histo and the VE PW histo so you can see where you are really lean or rich and at what PWs your errors are occuring. You can do the same thing using MAF histos. I like to use VE histos because I can see more data in more cells and for me it is easier to see the descrepencies. Lets say that my MAF is all good, but I am still having some rough issues. I can see by my VE AFR histo that I am running lean at 15kPa at 3600 and 3200 rpms, but I am fine at 21kPa at the same rpms. I look at the VE PW histo and see that I am running .671ms at 15kPa and .702ms at 21kPa. I look at my excell spreadsheet and see that .671 is closer the .250 adder cell and .702 is closer to the .375 adder cell. I know I need to bump my .250 adder up just a touch and I should be good. Keep in mind that these are not small errors we see due to adders. A small jump can turn a VE cell from a 17 AFR to a 13 AFR like *SNAP* that.

Did I miss anything?