About to play with quantity adjuster... any advice?

[Sausage]

if you want to mess about with the sdi map then use vagsuite not edcsuite: https://vperformance.co.uk/how-to-remap-diy/vagsuite-download-free/

 

 

 

Edited January 26, 2018 by Sausage

[mk2]

I will definitely try that once I've figured out what the cause of my problems are. Basically, I think that the SDI is capable of much more torque, with a remap. Not sure the box would take it though unless it's just been rebuilt.

I've also found out that the 1.9 SDI pump is exactly the same as the 1.7 pump. Well ok, there are a couple of  minor improvements in later 1.9 pumps, but other than some friction washers, the cam profiles are the same, the plunger piston assembly is the same, advance mech is the same, the operating pressure is the same etc etc.

I found this site that itemises all the internal components with bosch part numbers of both the 1.7 and 1.9... rather useful IMHO.

Lupo/Arosa SDI pump:

http://dieselcatalog.online/en/bosch/0460/0460404967.html

Golf/Leon and others SDI pump:

http://dieselcatalog.online/en/bosch/0460/0460404972.html

It is really bugging me why the computer (ECU) says "no" to more fuel. Mechanically, I can't find anything wrong other than a slight imbalance between injector quantities. Arrggghhhh!

[carrera-gt]

If nothings wrong with the pump and ECU says no more fuel, then either ECU has a problem or one of the sensors feeding data to the ECU has a problem.

If it's a fly by wire system is the correct deamnd signal reacing the ECU from the trottle position sensor ?

If its a cable system there will be a sensor somewhere that feeds data to the ECU.

If you have a spare ECU might be worth a try, on GTI's and similar you can check the output of the throttle position sensor perhaps you can do that here

 

[mk2]

Completely 100% agree with your thoughts...

I'm pretty confident that the ECU control output signal, that powers the coil inside the pump, which adjusts how much fuel to inject (quantity adjuster) is working ok. Why? When i pulled the top off the pump to check inside to see if it was jammed or something was stopping it from going to maximum fuelling, I watched the pickup arm briefly go all the way from minimum stop to max stop (which means it can...). I get the impression that the ECU calibrates the pickup arm and QA motion limits every time you switch on the ignition. When I played with the pickup calibration loop, I had to switch off and on again to get VCDS to show me the new limits (like 0.62 to 4.38 volts). So I'd say the control output to the pump is ok. It also means that the info from the pickup is also ok... The fuel temp sensor is also ok, as the data i get from it generally looks 'plausable'.

Drive by wire throttle pedal... Hmm. The only way I can test it is by looking at the live numbers I get from VCDS. No foot on pedal- reads 0. Foot to floor -100%. I can only guess that it's ok?

Temp sensors (air & coolant)- when cold, all sensors including fuel all read the same within a degree or so. Hot engine- coolant is around where it should be 90ish and air between 10 and 30 depending how long the car has been stationary (in this winter weather). So i'd say they're ok.

Needle lift sensor- tried two sets of (known to be good) injectors. Both report injection timing to be about the same- see above VCDS pic. So I'm fairly sure that is ok.

Engine starts, so crank sensor is ok.

EGR has been deleted, so that doesn't come into play...

I just Dunno...???

[carrera-gt]

Looks like the way forward is to measure the voltages directly with a volt meter not via VDS because that's what the ECU thinks and that may not be reality. if you check the wiring diagram there will be junction points in the loom where there are plugs where you can tap into the signals.

Perhaps two voltmeters would be good one to look at the trottle demand voltage and another to look at the pump drive signals while your drving it, after all your problem might simply be a bad earth on one of the signals.

[mk2]

I'm now at the point (really fed up), to try another pump... if i can get one cheap. I'll give a 1.9 pump a try as they're cheaper. I have a hunch that there is a fault with the plunger head assembly. 

I'm 99.9% sure that all the sensors or inputs are working fine. Before i do venture into mechanical territory, I'll have one last try at making logs of what's going on. I'm going to try and make a log of pulling away in 2nd or 3rd, uphill to the point where the car simply can't make it up to the top and stalls (it's not that steep either). The ECU should see a demand of 100% throttle. RPMs should steadily rise from around 1000 to perhaps 3500 and the down again to zero. The full power demand map should be selected by the ECU, with constant input temperatures, so the only variable will be rpm input with fuel and ignition timing outputs. If the injection quantity logs follow where they should be, then i'll know for sure it's the pump. The ECU has no way of monitoring actual fuel delivered, emissions or applied load, so i don't think it could ever throw a fault with the situation the car is in.

Measuring throttle pedal output I don't think is neccessary with a voltmeter (there are two potentiometers working in reverse to each other for safety). If the ECU sees 100%, then that is good enough even if they're not working right... I have the same theory with the pump servo pickup arm. Since it calibrates itself every time you power on (and you can physically see full scale motion of the quantity adjuster head), I don't think a fault there will matter either. It's a pretty smart system... however, if just after calibration you threw a switch and a potential divider in the feedback circuit, you could make the quantity adjuster servo move more (more power!). I might try that once fixed...

[Rich]

why don't you swap the sensors from the bad one to the working one and see which one stops it from being good?

[mk2]

First thing i did a few weeks back...

 

[mk2]

Got a log file for you experts out there...!

Here's a chart (dunno how to plot a line graph in excel with lots of data columns). You can see looking at the log that the ECU has decided to limit the max fuel delivery to about 23milli grams per stroke. I dont know why. It should keep going up all the way to 51mg just before you hit rev limit. So the ECU knows that it isn't throwing lots of fuel into each charge. It has in it's wisdom decided to limit the amount to just above a fast idle. Pedal or load is 100% (foot to floor). This was in 1st gear. Couldn't do the experiment in any other gear as I couldn't get the revs above about 1300 going steeply uphill.

It is almost like how a petrol car behaves when in limp mode. But there are no faults flagged up.

I don't think it's the pump that is the problem anymore. I could be wrong, but it does seem to me now that the ECU is limiting the amount of fuel being delivered which is why the car is completely underpowered.

[Pete]

Have you tried another ecu?

[mk2]

Not yet. Never done one on a Lupo... I'd have to mess about with the immobiliser lockout i think? I get the impression from the numbers that it looks like the ecu is following a map that restricts the fuelling (correctly). But the wierd thing, is why is the ECU following a restricted map. If i could get it to follow a 'normal' map, normal play would resume. Really odd.

[mk2]

Another random thought. Sorry to keep going on about this... how does the brake pedal switch interact with the ECU? What I mean, is what does the ecu decide to do when it sees the pedal pressed, and why? You can't do left foot braking with the sdi- tried that ages ago- the ecu goes into some sort of low power mode. Now, I wonder.....

[Pete]

I imagine it’s a safety override as well as an economy feature, so when the brake pedal is pressed it shuts off throttle inputs a - if you’re braking you’re not wanting to gain speed so fueling is cut to a minimum to save using fuel, and b - if you panic and catch both pedals when stomping on the brakes, it stops you going full throttle into whatever it is you’re trying to avoid hitting.

[Sausage]

Your Torque Limiter map is the likely reason. They just chopped it off at 23.3mg.

As said earlier your engine is atmo, you would need a turbo to utilise anything much over 27mg without clouds of back smoke.

You could edit the TL map something like this save it and try it (at your own risk):

There may be other maps limiting it like SOI so it may make little difference but worth a try.

Make the 800 mbar line a little lower than 27mg though or it will smoke when pressure drops.

Edited January 30, 2018 by Sausage

[Sausage]

Check the timing on your chart against the SOI map:

You can see they limit SOI as well above 24mg, so would need to add timing in the 25mg and 28mg columns a little for more fuel too. This would need to be calculated.

Edited January 30, 2018 by Sausage
words

[Sausage]

So it could be that you are just reading what the ecu is saying and expecting to be happening but in reality something is mechanically wrong with timing and / or fuelling (and compression) if you are down on expected power.

Edited January 30, 2018 by Sausage

[Sausage]

I would also be eliminating factors external to the engine as well like blocked exhaust, cat, excessive mechanical drag, differential, gearbox etc. Fuel pickup blocked, excess air in the fuel (which there will be if the tank strainer is blocked).

Have you found your ticking noise yet?

edit:  check your camshaft and lifters too, make sure valves are doing their job properly.

 

 

Edited January 30, 2018 by Sausage
more words

[mk2]

44 minutes ago, Sausage said:

I would also be eliminating factors external to the engine as well like blocked exhaust, cat, excessive mechanical drag, differential, gearbox etc. Fuel pickup blocked, excess air in the fuel (which there will be if the tank strainer is blocked).

Have you found your ticking noise yet?

edit:  check your camshaft and lifters too, make sure valves are doing their job properly.

 

 

Yeah, mechanically and fuel everything looks good. 

Still ticking, but only when hot.

[Sausage]

Are the temps right in vcds? So it is running the right maps.

You need another pump for comparison now I think.

Edited January 30, 2018 by Sausage

[mk2]

3 hours ago, Sausage said:

So it could be that you are just reading what the ecu is saying and expecting to be happening but in reality something is mechanically wrong with timing and / or fuelling (and compression) if you are down on expected power.

Well that is the truth! I've often wondered if it was possible to read what sensors are actually saying, not just what the ECU is seeing. The trouble is, the way the ecu works is by converting an analogue measurement to digital. If the conversion process is messed up, there is no way of ever knowing. VCDS only displays the sensors as the ECU is reading them, not necessarily the reality of what they are doing. There's meant to be a precision voltage reference inside the ECU, but is does have a tollerance. All measurements are based against the reference. If it's a bit out you'll never know.

I think the timing is ok. The numbers pretty much match your MAP data precisely. The exhaust is pretty clean- no smoke at all. I guess a bit like driving with a very light right foot.

compression is ok

The problem is 100% certain fuelling. But where and why.....? The wierd thing is, if the maximum fuel is really meant to be 23mg, then my engine should be at full power. I'd say i have about 25% power. The idle numbers are right, so the servo (quantity adjuster) is working ok. The output is proportional to voltage- or at least that is what the ecu is saying. Hmmm, maybe it's telling me injected quantities that are not true...

[mk2]

12 minutes ago, Sausage said:

Are the temps right in vcds? So it is running the right maps.

You need another pump for comparison now I think.

Yeah all temps are perfect. I even changed the sensors (except the one in the pump, which is covered in black encapsulation resin). 

Today i bought another pump. I'll swap it when i have time and let you know my results.

[mk2]

Hey Sausage, there's an anomily in my log that makes no sense to me... if you look at the data in the two blue boxes, you'll see that the delivered fuel is the same, but the servo position measurement (voltage) is different. The way I understand how the quantity adjuster works means that is impossible. What the??? If the voltage goes up, it means that the physical position of the arm is close to max. Which means that the dump ring is shifted over to the right = more fuel. Lower voltage means the arm is closer to min. 0.6v means no fuel. 4.5v would mean max fuel. Why is the 'calculated' fuel delivered per stroke the same? If the start of injection is earlier, the duration of the 'squirt' is longer because the 'dump ring' stays in the same relative position. If the QA moves to the right and injection starts earlier, the amount of fuel per squirt 'has' to go up. 'Has to'. The numbers are complete nonsense!

[Sausage]

You run out of SOI at 24mg. Without more duration you cant get the fuel in. Also your TL is 23.3mg.

As said above it's likely what you are measuring is just what the ecu is doing and expecting from the maps not real world figures, it has no way to know what is really happening after all beyond certain factors.

 

[Sausage]

Missed this bit, ticking noise when hot = hydraulic lifters as prime suspect.

This sounds like fuel starvation, blocked fuel filter, blocked tank strainer, excess air in the fuel. Try clear pipes between filter and pump.

Edited January 31, 2018 by Sausage

[mk2]

11 hours ago, Sausage said:

You run out of SOI at 24mg. Without more duration you cant get the fuel in. Also your TL is 23.3mg.

As said above it's likely what you are measuring is just what the ecu is doing and expecting from the maps not real world figures, it has no way to know what is really happening after all beyond certain factors.

 

I guess that does explain why the only way to up power to get more fuel in a shorter time (high rpms) is to change the injectors to bigger nozzles... I know this to be true for direct injection petrol engines I've done ECU design work on. At high revs, you simply run out of injection time. But by changing injectirs to a bigger flow rate has consequences at idle. It is much more difficult to meter fuel at idle with big injectors.

I think there is a fault on the plunger. Not sure exactly what, but it's almost behaving as if there's a slight taper or cone on the shaft, which changes the fuelling profile from linear to logarithmic. So instead of (for example) 1.0volt giving 10mg and 2.0volts giving 20mg, i get 1volt 10mg and 2volts only 12mg.... So even if i correctly set the idle quantity adjustment to 2.8mg, at high throttle demand i don't get sufficient fuel, even though the ECU thinks i do. Or at least that's my current theory! It does explain why if i set idle at 2000rpm, i get huge power at full throttle- i've physically moved the dump ring to a new unaffected part of the diesel plunger piston.