Injection Volume Math needed

Hello!

I've started fooling arround with Vagsuite and MPPS. Don't worry, this car is a test car meant the be broken

I ended up wondering how to calculate the maximum Mg/Stroke possible, as there must be a limit.

I guess timing, nozzle size, fuel pressure and plunger piston's size are related, but I don't know much about how to estimate the maximum delivery possible.

Let's say I've got an auto jetta ALH (11mm pump, 0.158 nozzle), how could I find out the limit? I'd like to learn the math behind it, so I can apply it to any nozzle/pump combos.

You need an injection pump test stand or a calibrated injector flow test bench to get that info empirically unless you are using stock pump injectors. For stock injectors the info is already in your ecu- at least in the range that the factory deemed necessary to calibrate. (hint: the full "map" may not accurately reflect reality! Little known fact) For an ALH 5 volts is as much fuel as you can get (actually it usually a little less depending on where its hammered) and the IQ/volts relationship is not linear across the full range.

So in your case essentially just try it and see how far you can push it safely with trial and error.

While you're experimenting, plan on getting an EGT gauge because injection duration will get longer with more fuel and stock nozzles.

The performance of a pump is usually measured in cc/1000 shots. This is done on a test bench with calibrated nozzles. Contributing factors to overall performance are plunger diameter and throw. The throw of the plunger is determined by the cam plate used. The limiting detail that you need to be aware of is the ability of the pump to recover at a given rpm. That is to say you don't want to run the pump dry because it can't get fuel back into it fast enough to replace that which was pumped out.
Four cylinder Cummins ve pumps of the 12mm variety can provide enough fuel to support 300-350hp at well under the 5k rpm that TDI engines can rev to. By comparison the stock nozzles have approximately the same orifice size as race 520 nozzles, delivering fuel to an engine that is 2 litres larger.
Basically, what I'm driving at here is that the pump is not going to be a limiting factor. If you want to scientifically approach and qualify your experimentation with tuning your IQ, you can purchase a wideband probe and gage set to measure your results (17:1 is a good ratio to shoot for, iirc). That, along with a pyrometer and vcds should provide all the tools that you need to optimally tune that beast.

Sent from my mobile look-at device

Hello, thanks for your replies.

These bring me more questions though haha

Does the engine see "actual" volume (mg/strk)?

or

Does the engine see an "estimated" flow; as fueling must be determined by a voltage range (0-5V?), higher meaning more I guess, and translates it to 0-51 mg ? Therefore, having bigger nozzles would deliver more "actual" fuel per injection, but the EDC still thinks he's injecting like 51mg at max voltage, but in fact it's injecting more? Meaning that the tune should LOWER IQ (or increase air flow/boost) to get smoke free and low EGT ?

Knuckle720 said:

Hello, thanks for your replies.

These bring me more questions though haha

Does the engine see "actual" volume (mg/strk)?

or

Does the engine see an "estimated" flow; as fueling must be determined by a voltage range (0-5V?), higher meaning more I guess, and translates it to 0-51 mg ? Therefore, having bigger nozzles would deliver more "actual" fuel per injection, but the EDC still thinks he's injecting like 51mg at max voltage, but in fact it's injecting more? Meaning that the tune should LOWER IQ (or increase air flow/boost) to get smoke free and low EGT ?

Click to expand...

There is NO measurement of fuel in any TDI...

Your car uses crank acceleration to calculate a value that is factored into the injection pumps quantity adjuster position voltage value.

Knuckle720 said:

Hello!

I've started fooling arround with Vagsuite and MPPS. Don't worry, this car is a test car meant the be broken

I ended up wondering how to calculate the maximum Mg/Stroke possible, as there must be a limit.

I guess timing, nozzle size, fuel pressure and plunger piston's size are related, but I don't know much about how to estimate the maximum delivery possible.

Let's say I've got an auto jetta ALH (11mm pump, 0.158 nozzle), how could I find out the limit? I'd like to learn the math behind it, so I can apply it to any nozzle/pump combos.

Click to expand...

Simple answer, too many variables to generate any meaningful data "On the car".

To calculate Pump Volume is one completely different set of test perameters.

Pump installed and run at a FIXED rpm and FIXED temperature (40C) using a Calibration fluid with a precise viscosity (ISO 4113 / 1478-AW2 -> Bosch Requirement).

Testing a pump requires using a test orrifice often referred to as a "Test Injector" these are set to flow very specific volumes and are needed to generate the prerequisite restriction to generate the peak pressure by the pump.

Test injectors are mandatory in order to measure the pumps high stage leak-back volumes, these are used to determine the pumps condition. Without knowing your pumps internal leak-back data, it is literally impossible to calculate ANY injection volume from your pump.

FYI, notice we haven't even gotten into the injectors yet?

Generating laboratory data on the vehicle are virtually impossible, thats just the fact of the matter.

Laboratory data that is used to build the ECU models is not relevant to in vehicle use, that "perfect" data is used to determine the control ranges and then variables are determined thru in vehicle testing to see just how much "Slop" needs to be accounted for.

The easiest way is to just set your pump to 2.8idle IQ, middle of the range timing, and put it on the dyno and measure engine output. Thats the best you can do aside from removing and individually testing the components in a calibrated fixture using close tolerance fluids and all within very precise operating temperatures and conditions.

https://www.youtube.com/watch?v=29lpNDA-5MI

Hello! Thanks for your replies. I'm now wondering:

If the computer doesn't see actual IQ, this means a tune could/should use a "near stock" range? Like 0-51mg ? As if 0 and 51 desired IQ could be compared to maximum/minimum pump voltage?

Ehhh..

51mg value doesn't reach 5.0V in stock tune, so many times it's necessary to raise the voltage definition values upwards proportionally. But all while test and tuning.

But it's not really a good idea to run the pump's maximum fueling at 5.0V.

Knuckle, there are lots of guys that are not professional tuners that will give away copies of their handiwork. What I really mean is get your hands on a Stage 1 or so tune and go through the maps and see what was done; if the numbers look reasonable to you you can modify a copy of your original tune to look the same and flash it on and run logs. It is not always a good idea to find a similar model and year and assume you can flash it on w/o problem; you might just brick your ECU and then it is a real pain to recover it back to "normal"... Just a thought.

Check out this thread, this guy may be of some help to you: http://forums.tdiclub.com/showthread.php?t=478229

As Nevada_TDI says, I based my work from the topic that is somewhere on this forum, and from a couple guides about EDC15 that are on ecuconnections, then I went and download around 20 rate my tune files for ALH, AHF and ASV engines, while also reading the comments about what was good and bad about the tunes.

The best one/best well explained one is for sure hneel work:
http://www.ecuconnections.com/forum/viewtopic.php?f=152&t=11343

I also recomend you to read the EDC15_rev1.3.pdf(also available on this forum in some big thread about ecu tunning that I never remember the name ):
http://www.ecuconnections.com/forum/viewtopic.php?f=2&t=867

Then a cheap MPPS cable, always remove the number 15 fuse and connect a battery charger so you dont end up with a bricked ECU due to voltage sag during the write.

Today I pulled all my maps into a calc/work sheet(aka excell) and revisited all the maps, bumped the torque limiter a bit on the low and high end, and will try a new smoke map down to 17.8:1 AFR, VagEdit/EDC15 Suite and see if I kill my clutch and or end with a smoke train and revert back.

Its not as easy as paying for a ready made tune, but its all that hard, that being said I have part of a masters in electronics engineering and my work is grabbing and idea an making a schematic, laying out a pcb, programming if need, making pc interfaces, designing enclosures in CAD, put them to CAM, machine and deliver a ready made product, also, repair laptops on the weekends, and have been tunning two strokes since I was 13, I never paid for something that I could learn, I will always learn how to do it and do it my self, not because I'm a cheapskate but because I like to understand why and how things are done and how they work.

EDIT:

Regarding the maximum injection, your pump/nozzles are already setup to deliver 51mg/str, but your stock maps wont ever go near those values, my codeblocks for automatic specify a maximum of 36mg/str of IQ, there are still 15mg of fuel that can be delivered that are not being used by the stock maps.

If you want to start playing around, get VagEDCSuite(version 1.3.9), create a project, load your dumped bin, and just look around all the maps, a basic tune will involve raising SVBL, raise boost target, raise the 85 and 100% drivers wish a bit so its always the higher value, raise your torque limiter a bit, and them export the smoke limiter map to excel, just divide the IQ values by the axis MAF air mg/str and you have AFR, as usual the original map wont go over 800mg/str of air, so you change a couple axis values to go up to 950mg/str and you change the values to stay above 18:1 and with a maximum of 2100/1.1 bars of boost you are in the 100hp territory, also, fill the EGR map with 850's so you delete it.

EDIT2: Lots of typos/bad grammar I know.

Odd, MPPS told me to pull fuse #11 to kill the cluster, but my car isn't a Euro version.
Senso, thank you for speaking up and sharing the "rate my tune" idea, there needs to be a lot more of that. I am thinking about playing with some Stage 1 type ALH tunes for a couple of friends and their stock MK4's. I am not teaching summer school this year, so I may learn to tune this summer just to pass the time; the time that doesn't include tons of yard work and well needed house maintenance.

Its either #11 or #15, I pulled the number 15 and it stopped complaining after I tried doing reads in a row just to test how reliable was the connection, also, my car was imported from Canada by its previous owner, so it as little unusual things like blue tinted mirrors and the oh so funny objects in the mirror are closer than they appear, that one was funny to read, got my mirrors labelled as Jurassic park props lol.

If you want I can leave my maps here, or create a new post explaining what I did..

I can also give you a little help and change a file and explain you what I did.
By no means I'm a pro in this, wont even consider myself an amateur so quickly, but I'm eager to learn and to also teach what I have learned so far.

I think that being a newbie around here and in other ecu related forums people wont take credit/take their time to respond me, because I might be another troll and/or annoying kid asking for a free tune..

But, so far it seems pretty straightforward on how to do a stage1 tune, and for the time it takes to learn and do it, for me is pretty worth it, companies/random facebook "companies" charge from 150 to 500€ to do a stage1 tune and most of them will end in a smoky tune because people in Portugal LOVE smoke trains..

On a 2003 I don't pull either fuse. I have had to pull fuses on earlier cars.

My cable will ID the ECU and do one read(never tried a write without puling the fuse) but them wont talk to the ECU unless I cycle the key to OFF and ON again, with the #15 it will just work, so better safe than sorry.

Also, dont forget about the K line that goes to the radio, I have an aftermarket on and I bought a pigtail that plugs into the original connector and then gives me two din connectors common in aftermarket radios and the K line is isolated, but if you can't talk to your ECU pulling the stereo is something that should be tried as an earlier troubleshooting method.

I find the use of pirated software and cheap chinese ripoff cables extremely objectionable. But I understand that some people also like to install their own toilets or paint their own walls or pour their own concrete, and others would rather hire a plumber or painter or mason to make sure it's done right with assurance and minimum hassle. I can (and have) done my own plumbing stuff but personally I usually just pay a plumber to do it, becuase there are better more productive uses of my time.

Just a couple of points I would like to offer.

senso said:

Regarding the maximum injection, your pump/nozzles are already setup to deliver 51mg/str, but your stock maps wont ever go near those values, my codeblocks for automatic specify a maximum of 36mg/str of IQ, there are still 15mg of fuel that can be delivered that are not being used by the stock maps.

Click to expand...

got to be careful with assumptions like this. For example take a look at an SDI file. Same injectors, same pump, but the volts map "stops" at some value just over 25 mg or so. In other words, when Bosch provides the calibrations for the specific application, they will often only calibrate the maps for the needed range or the anticipated operating range. The rest is just filler values. This happens a lot. Not just in EDC15.

And keep in mind that if you change the pump volts at all, you are pretty much invalidating whatever calibration that existed and making your own. I see a lot of these neuvo experts trying to look smart by criticizing "pro" tunes for being "incorrect" and "not extending the range of the axis" or whatever. What they fail to realize is that the only way to be "accurate" is with a physical injection pump test stand to actually experimentally measure the flow of a given pump/injector combo. Anything else is usually just a guess "that looks nice". What really matters, is where your pump volts are in relation to 4.9xx volts which is full scale (full fuel delivery, full travel stop on the positioner) and idle (which is the minimum fuel flow reference volts). You can call 4.9xx volts 45 mg, you can call it 25 mg, you can call it 51 mg. What matters is where (the applicable) limiting torque relates to throttle position and all the other parameters that take an input from the (applicable) torque message inside the ecu. (it's not going to be "accurate" anyway once you start tweaking it, but that doesn't mean you can't still do a good tune) As long as you understand what changing that relationship does to all the other things that matter, you are ok. It gets more complicated because various (hundreds of) parameters may take their instructions from different torque messages. Knowing these subtleties is part of what separates the men from the boys- and I'm pretty sure 99% of tuners don't know or care, which ultimately affects the efficacy of the end product. It's still possible to change maps and get a result, but that's more like part of a tune. It's fine for tinkering and figuring things out and experimenting, but there is a lot more that goes into a comprehensive ecu recalibration (aka tune) than you can see with EDCsuite.

senso said:

and them export the smoke limiter map to excel, just divide the IQ values by the axis MAF air mg/str and you have AFR, as usual the original map wont go over 800mg/str of air, so you change a couple axis values to go up to 950mg/str and you change the values to stay above 18:1 and with a maximum of 2100/1.1 bars of boost you are in the 100hp territory,

Click to expand...

For stock injectors you don't need to mess with that as much. For the basics, pretty much just get the smoke limiter to be higher than the torque limiter at full throttle (only) and you are good. More or less just get the smoke limiter "out of the way" so the Patm torque limiter wins at WOT. (another common mistake I see even with "pro" tunes, but that's a matter for a different discussion) Smoke map is already calibrated correctly for stock injectors, just need a bit more headroom when you tune. Stock injectors in a healthy ALH won't smoke even with the positioner pegged at 5.0 volts (barring warp field collapse) if you add a couple psi boost. If you have big injectors, now that's an entirely different matter. There's a bit of art involved in getting the smoke/response balance to suit your personal preferences.

Thanks for you input Kerma.

I didn't touch pump voltage at all, and yes, there are differences in pump voltage maps in engines with same injectors and pumps. I know, its an hydraulic system there are some injector bodies that will work at 180 or 190 or 200 or 210 bars and that will affect how the injectors will open/work.

That and pumping losses and whatnot, if the original ALH pump map only goes to 40 or 41 is because above that, there be unicorns, the original maps dont request that much so why bother.

That and if I touch the pump voltage I would be literally pulling values out my rear end, in the future I would like to get some bigger injectors, and after that all bets are off, the best way to get it semi-accurate is to search for and engine code that used that same size injectors and copy the map, that or make a jig out of a brushless motor an ESC and a canibalized wiring harness and program a micro-controller to exercise the plunger through its whole range at each rpm point, make the injectors squirt X times into a graduated flask, seems like a nice way to spend a day or two.

I have done something kinda similar to map and test ignition systems for the two strokes(I also have developed a digital ignition CDI, nothing even as complex as the EDC15, but its a start).

Also, of course I'm not saying I'm a pro tuner, and I know that EDCSuite is not the be all end all, I have poked at it with HxD, but I wont use WinOLS, because, I cant afford it, from the info available I can pull a map with an hex editor and massage it in excell and then put it back into the bin.

And yes, I know that using cloned cables is very bad practice, but I cant afford a 2k € cable, if I had 2k for a cable I wouldn't have bought the that that I have today.
The most honest way would be to make a cable from the scratch and read the EDC15 manual to understand the programming sequence, or rip the ECU open and read the flash/eeprom with something like a TL866 or even a simple programmer with an Arduino and a couple shift-regs like I made to dump old 68K rom's.

There are some/ a lot of ALH's smoking like trains in Portugal, they might not be that healthy, but its possible, heck, people pay to make them smoke.

Sorry in advance for any barbarity that I might have written.