Jump to content
nilsp

short pulse with adder table

Recommended Posts

Hi,

I'm new to this forum.

 

I'm preparing to start my e30 m3 s14 engine with g4+

I'm going to use sequential injection and have som new bosch injectors and want to prepare the short pulse with adder table.

As far as I understand I can only adjust up to 4ms?

Have got some data from motec on these, and they seem to need adjusting the table up to around 7ms.

-made a sheet in excel..

 

Is this possible or are there something wrong here!?

 

Thanks!

Nils

bosch dyser korreskjson.xlsx

Share this post


Link to post
Share on other sites

That is not a short pulse width adder table.  Typically short pulse width adders will only be a few 10ths of a mS.  You couldnt really calculate short pulse width adders from the attached data since you will need many more measurement points between 0 & 3ms.

Share this post


Link to post
Share on other sites

What injectors are they?

Out of interest (and seeking data for my own injectors, although yet to use them with the Link so can't confirm if this works), in the past I took the data from Motec M1 Tune (which is what Nils' data looks to be) for my own injectors and put it into a spreadsheet and did some maths. BTW, I don't think the columns are labelled right - eg in 10v I think it should be 1.00ul of fuel requires 1.636ms of injector time.

  • You can work out the dead time implied for each point given Motec's assumed reference flow (and pressure)
  • At 10v - 1ul delivered in 1.636ms, should take [X]ms (depends on injector size) - difference between 1.636 and X is deadtime
  • Repeat for whole column
  • I then averaged the deadtime once it became more or less linear (about 8-9ul), and called the difference where non-linear the "short pulse width adder"

I tried with the data ID supplies for ID1000s (for use with M1 Tune), and the deadtime using my method is a very close match to the deadtime they show for GM characterisation. The short pulse width adder less so (eg for effective pulsewidth of 0.25ms, my maths says to have a short pulse width adder of 0.075ms, whereas ID  would add 0.134ms. 

Now I don't know if the above is accurate but it seems reasonable... use at your own risk, etc.

Interestingly Motec's data for ID1000s is different to ID's data....

Edited by CamB

Share this post


Link to post
Share on other sites

 For what its worth, in the version below I have added the maths I think is right (shows deadtime pretty similar to the M800 style deadtime, and a mathematical attempt to work out what the short pulse width adder would need to be). Happy to be corrected if I haven't worked it out correctly - I am here to learn!

0 280 158 123 Customer sheet v2.xlsx

Edited by CamB
Changed spreadsheet to add a better spreadsheet

Share this post


Link to post
Share on other sites

Hi,

 

Thanks a lot for your effort,really appreciate it!

Before I saw your answer I tried to work a little in an spreadsheet myself, and got a similar graph.

-But I was a little unsure about where to put the graph against the PW ms times.

Because on the spec sheet it says minimum 1 ul which is at a little over 1 ms with 14v.

Do I get something wrong regarding the deadtimes then? I guess so :)

-I then thought the minimum PW also was 1ms?

 

BTW,already updated my adder table with your numbers, hopefully soon ready to start the engine.

I use the same ecu on another engine now,with smaller injectors and batch fire, but haven't noticed or thought of the PW times,so have no idea of what's normal times here!

 

Thanks again!

Share this post


Link to post
Share on other sites

I'm learning as much as anything through this process too.

This is the best discussion I am aware of:

http://injectordynamics.com/articles/gm-injector-characterization/

The minimum delivery of 1ul is an effective pulse width of <0.1 msec for your injectors (1/10.72). (would be great if someone checked my maths - I assume if the injector is 10.72ml/s this is  10.72ul/ms)

Do you mean how does this relate to the Link setting "minimum effective pulse width"? Hopefully someone at Link reads this but my working assumption based on what the help file says is that this has two jobs - (1) to set the minimum recognising that injectors won't open at all below a threshold and (2) to stand in place of accurate short pulse width adders. For example, if my spreadsheet is correct (not guaranteed!) then the minimum could be 0.1 ms, but only if the short pulse width adder was right. If you had zero in the short pulse width adder table you'd want to have closer to 0.6ms (to account for the non-linear injector response down there).

I guess if I was in your position I would put something conservative in there, like 0.2-0.3ms. Then if this minimum is kicking in at idle (which I am not good enough at maths to work out), based on logging, try lower and see if it still runs. Would be interested in others thoughts?

Edited by CamB

Share this post


Link to post
Share on other sites

 

Do you mean how does this relate to the Link setting "minimum effective pulse width"? Hopefully someone at Link reads this but my working assumption based on what the help file says is that this has two jobs - (1) to set the minimum recognising that injectors won't open at all below a threshold and (2) to stand in place of accurate short pulse width adders. For example, if my spreadsheet is correct (not guaranteed!) then the minimum could be 0.1 ms, but only if the short pulse width adder was right. If you had zero in the short pulse width adder table you'd want to have closer to 0.6ms (to account for the non-linear injector response down there).

I guess if I was in your position I would put something conservative in there, like 0.2-0.3ms. Then if this minimum is kicking in at idle (which I am not good enough at maths to work out), based on logging, try lower and see if it still runs. Would be interested in others thoughts?

The "minimum effective pulse width" setting becomes important when you have reasonably large injectors and dont have proper short pulse width adder data.  You can use this setting to stop the ecu from commanding a pulse width that is below the injectors linear or predictable response area.  So for instance if you know your injector flow rate is not linear "falls of the cliff" below 1.3ms pulse width then you set the "minimum effective pulse width" to 1.3ms and that will prevent the ECU from ever commanding less than 1.3ms even if the calculated fuel flow requires less than that.

Of course if you have short pulse width adder data then you are better to have a much lower setting in "minimum effective pulswidth" to allow that table to do it's job.  Having said that it is not common that you see engines running sequential that need pulswidths less than about 0.8ms at idle so your short pulse width adder table often will not need to go much below that point.

Share this post


Link to post
Share on other sites

Just a little update here, been busy with work and getting the engine run in dyno etc.

 

But I used the above settings for dead time etc, and the engine has run very smooth.

 

Regarding the minimum effective pulsewith, I didnt really remember to take a look and dont have a log to check now, but think it was somewhere between 2-3ms, so didnt really turn out to be a problem.

 

Only strange thing is that the "modelled" fuel tables were getting very high numbers, were up to over 140 at one spot, the max value is 150.

Something could be wrong in the setup, but not sure what!

Share this post


Link to post
Share on other sites

Regarding the minimum effective pulsewith, I didnt really remember to take a look and dont have a log to check now, but think it was somewhere between 2-3ms, so didnt really turn out to be a problem.

 

That doesnt sound right at all.  This number should normally be set somewhere around where your injectors start to behave erratically, usually this will be <1.5ms. 

 

Only strange thing is that the "modelled" fuel tables were getting very high numbers, were up to over 140 at one spot, the max value is 150.

Something could be wrong in the setup, but not sure what!

This means there is more fuel going in than the ECU is calculating.  It could be many things, but most likely one of your injector settings such as flow rate or rated pressure, actual fuel pressure is greater than you think, or perhaps something like MAP sensor calibration is incorrect.  Since you appear to have 30-40% more VE than expected you have something fairly wrong.

Share this post


Link to post
Share on other sites

Hi Adam,

 

thanks for reply.

 

havent had opportunity to check the pulsewith yet.

Engine was in dyno, now removed, awaiting install in my car.

 

Regarding the VE numbers;

what are normal numbers for a good engine?

We checked the injectors flow because of the high numbers,and it was looking to run leaner and leaner,so suspected injectors could the restricted or something.

-they were not, but at 3 bar they gave around 550cc, should be 620 @ 3 bars. Also the actual fuel pressure is more like 2.7-2.8

The ecu is programmed to 620cc and 3 bars.

 

Guess this could explain some then? 

 

Brgds,

Nils

Share this post


Link to post
Share on other sites

I would recommend you install a fuel pressure sensor into the ECU. There not that expensive, also gives you valuable info that can saves hours of fault finding. 

Share this post


Link to post
Share on other sites

The fuel pressure was logged through the dyno, not the ecu.

But the ecu thinks the pressure is 3 bar then ofcourse,since thats what the setting is set at,and the equation is out from.

My injectors should be more than big enough,and I'm running stock FPR.

 

Could install a sensor ofcourse, but I feel its working ok for now..

Running wbo2 all the time and have big fuel pumps, so think the supply is under control.

M3 S14 2.66l NA making 315 hp and 325nm, and just broken in.

Share this post


Link to post
Share on other sites

If I have my maths right, the correct fuel pressure difference might reduce your VE numbers by 4-5% and the difference in injector flow by 10-11%.

 

Share this post


Link to post
Share on other sites

For what its worth I have been working on my own spreadsheet and incorporated it into the one above (v2). I think its a better calculation working off the ability to nominate which portion of the data represents linear flow, and it interpolates for the short pulse table.

For nislp - I wouldn't suggest you change anything if you are already tuned. The difference is minor and you'd have to retune. 

0_280_158_123_Customer_sheet_v2.xlsx

Edited by CamB

Share this post


Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...

×
×
  • Create New...