Jump to content
mk2

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

Recommended Posts

Have just changed the injectors over and have cleaned everything inside the pump. New seals and everything is god to go. I was about to start messing with the QA settings (everything is still factory- but a bit out), but thought I'd see if anyone on here has ever been brave enough to get stuck in. I know you can change the injection start point, but the servo mech in the pump can be 'tweaked'.... Can't believe how much metal material was stuck to the magnet  (for the servo motor arm).

On the same subject, does anyone know how the adaption routine works?

Share this post


Link to post
Share on other sites

Well I've nudged the servo assembly over to the right and now we're idling at about 2000 rpm. Huge increase in power/torque. So I need to drop the revs using Skezza's adjusting thing... Will keep you posted.

Share this post


Link to post
Share on other sites

On a side note i have the SDI map and it's got all the maps needed for a turbo just sitting there, so shouldnt be too hard to turbo one. pump adjustment and voltage maps etc too obviously.

Share this post


Link to post
Share on other sites

I can't think how the other fuelling maps could be accessed on a SDI. I mean the only inputs are rpm, temps, pedal and injection begin/end. Looking at how the pump works, it doesn't take a rocket scientist to realise that the amount of fuel to inject per stroke is primarily determined by how far to the right the 'dump ring' (as i call it) is shifted. More right, more fuel. Left, less fuel. The rev limiter is 100% electronic. If i had a spare pump, i'd be tempted to rig up a cycle brake cable or somthing a make a manual throttle as an experiment. The ecu would throw a wobbly, as the engine would be doing exactly what was demanded by the driver..... i'm amazed at how simple it is.

Share this post


Link to post
Share on other sites

What system do you use? Is it the unlocked version of vagcom (vcds)?

Share this post


Link to post
Share on other sites

Edcsuite probably easiest for looking at the map.

MPPS for upload / download.

Share this post


Link to post
Share on other sites

Ok, so I've tried to figure out if these settings are right (without any reference sources). See screen shot.

There's an adaption setting somewhere, which I believe can be changed to alter the sensitivity of the throttle servo- something like change 32768 to 32700? But where is it???

Right now the car is completely gutless, but the QA adjustment is right. I think the upper servo pickup reading is low however. Before I unscrew the top of the pump and adjust the sensor arm position, I'd like to try the adaption routine. Think it may do anything?

QA settings.jpg

Share this post


Link to post
Share on other sites

Pump volts and IQ look right.

pump volts.jpg

Copy paste from another forum for A4 tdi:

----------------------------------------------------------

Group 019 (Quantity adjuster)

Voltage from modulating piston displacement sensor: 0.500 - 0.970 V 
Voltage from piston displacement sensor: 4.150 - 4.740 V 
nothing
nothing

If specfied values are out of tolerance, possible malfunction of the injection pump quantity adjuster.

-----------------------------------------------------------

ideally you need the spec for your pump

 

Edited by Sausage

Share this post


Link to post
Share on other sites

That's a really useful plot. Awesome info!

Yeah the pump is the same across the entire range- i think the VE diesel pump is used for lots of makes and models. The only thing different physically are the addition of a turbo or different injectors. Our SDIs use 5 hole 0.170 size nozzles. The settings should be about the same as a 1.9 SDI or TDI, as at idle the engine drag is about the same. They all need about the same amount of fuel feed to keep the RPMs at about 920.

So maybe my measurements of 0.620V to 4.380V are ok after all. Before i shifted the servo mechanism to the left the readings were different, like 0.60 to 4.7. I had a lot more power, but the IQ was about 11 or 12, which is obviously wrong. The hammer mod is great, but it only tricks the position pickup sensor. I don't think the IQ quantities are measured at the no. 3 injector (even though though technically it would be possible)- the numbers shown in Vcds are calculated form the map. But i just don't know for sure. 

When i get time, i'm going to plug into my other SDI and get some readings (that car runs like brand new).

Do you know how adaption works, or where to change it?

Share this post


Link to post
Share on other sites

Just had a closer look at that map table... if you look at the max servo voltage, upper reading it measures 4.380 volts. On the map chart that equates to about 27 milligrams per stroke of fuel. The max fuel per stroke at top RPMs is about 51, which clearly explains why the car is gutless, running only about half power! To get maximum fuelling, that upper measurement 'has' to be at least 4.50 volts...

I'm not sure, but i think the only option is to adjust the servo pickup arm position, by moving it clockwise. Undo locknut, move then tighten up again and re-measure eveything.

Share this post


Link to post
Share on other sites

No not really, that is one of many maps that limit the engine, it will take the lowest of several maps under different loads and conditions.

Here is drivers wish (pedal %)

 

sdi DW.jpg

here's another Torque limit (to protect gearbox, clutch  etc)

 

sdi torque.jpg

Your engine is atmo so will only be able to handle around 27mg of fuel.

It may be that some of these maps are redundant anyway depending on how mechanical the pump control is, so just because there are maps for things doesnt mean they are doing anything on an SDI. (it has turbo maps after all in there)

Edited by Sausage

Share this post


Link to post
Share on other sites

Uh, ok. Any suggestions on how to get my power back to normal...? I'm sure its something to do with adaption, but i've never played wuth that on an SDI.

Share this post


Link to post
Share on other sites

Pump timing external and internal (if you rebuilt pump), pedal /TB, IQ adjustment? 

Here's the hammer mod but i assume you have read this or similar: http://www.myturbodiesel.com/wiki/injection-quantity-adjustment-fuel-pump-hammer-mod/

I assume it is similar on SDI pump.

Pedal / throttle body (dont know is SDI needs this): http://wiki.ross-tech.com/wiki/index.php/Throttle_Body_Alignment_(TBA)

Why have you done injectors and pump anyway? I assume a problem. You need a baseline before making changes otherwise you dont know what is wrong.

Share this post


Link to post
Share on other sites

So here's the history to explain all my madness... over the last year.

Engine sounded ok but occasionally when hot there was a distinct ticking or pinking sound, as if it was injecting too early. So after messing with pump timing (no effect), running a purge and changing all filters, came to the conclusion there must be a blockage in the tank. Engine was now down on power by a lot and ticking noise all the time (I have two SDIs so can compare easily!). So checked the strainer in the tank. No air leak in fuel lines either. All fine. Hmm.

Must be the injectors. So whipped them out and after giving them a good clean, took them to a local place that has an injector tester. All five ports on each injector just fine. All good there too. Chatted with the diesel specialist who thinks its dead cert that the pump has had it. After 92k miles?!?

So after much research things started to point to the quantity adjuster servo system being out of whack which can happen apparently (according to interweb).

Ok, so out with VCDS last week and sure enough the injected quantity reading per stroke was way too high at idle- like 10mg. Did the bump (hammer) mod thing (as it's called!?!)- or servo position adjustment as I call it, and nudged it down to 4mg. Drove cr and definitely an improvement, but still very much down on power. So as an experiment, nudged it right over to make it idle at 2000RPM when hot, and my god my SDi turned into a rocket ship. Way more power than stock. Ok, so IMHO the pump must be ok.

So, I conclude that there is a problem with the QA servo or something related to that. I'm guessing that the servo pickup is telling the ECU that the amount of fuel delivered is too high, so when I demand full power, it thinks I'm getting a full load of fuel, but in fact the pump is delivering probably less than half.

So this evening I whipped off the top of the servo head to see what s going on in there. All looks ok. Clean and no jams or anything. I watched how the ECU calibrates itself on powerup, to get the max, then min voltage limits, so it knows where the arm is sitting, in relation to the fuel dump ring on the pump shaft. So I thought if I can change those measurements, I should be able to shift the idle point and subsequently the max power point. There is a calibration reluctor ring on the left side of the horseshoe pickup which you can move to change the limits- see pic. Tried changing them to max then min, while doing the QA injection calibration to get idle back to 2.8mg, but still no change. I'm just about out of ideas!

So presently the idle setup is like below. I did notice that the injector variation is quite high which I'm a little concerned about- it might indeed mean my pump's knackered. other pic.

 

cal ring min max.jpg

variation.jpg

Share this post


Link to post
Share on other sites

Checked throttle response- as in does the ecu see it go from 0-100 - no problems there.

Also swapped out injector no1 for another known good one, to see if i can get the injection quantity balance evened out. If you look at the measurements above, no1 injection is being added to. By my calculations, if the ecu is being forced to add extra fuel for each injection pulse, it must mean that one of the holes is clogged. So i remeasured after swapping and going to work and back, and now it's even worse! Huh? What the carp is going on?

I now have a hunch that the plunger/distribution thing is leaking on cylinder one outlet. Basically pump is probably buggered. Anyone got a spare SDI pump that i could try?

 

Share this post


Link to post
Share on other sites

I just had another suggestion from another forum that it might be the shutoff valve N109. Fingers crossed that it is... anyone else experienced wierd SDI performance caused by N109? When it stops raining, I'm outside with a spanner.

Share this post


Link to post
Share on other sites

I dont think (but dont know) those variations are big enough to be a problem, they may indicate some slight difference in compression, injector wear etc but nothing significant. (as said i dont know, but thats what i think).

I would cover your bases by doing compression test, and make sure the cambelt tension and timing is correct before messing with pumps etc. Once you know that is all fine then you can know you are not wasting your time doing other things.

Ticking noise could be injectors but could also be piston rings for example. You need to narrow it down to head or block or pump etc with a stethoscope.

Edited by Sausage

Share this post


Link to post
Share on other sites

Wasn't N109 shutoff valve. Removed, examined, cleaned and replaced. Engine still very low on power. 

I haven't done a compression test, but i don't think it's a mechanical issue. When i moved the injection Quantity Adjuster to read 0, the engine idled at right up at 2000rpm, there was loads of power at full throttle- more than stock. I'm now beginning to think that the plunger/distribution shaft inside the diesel injection pump has a wear scratch/scuff on it at low/medium fuelling position. If i set the QA to be correct, then it idles perfectly, but i can't get decent fuelling at 100% throttle (even though the ECU thinks it's ok). 

My first thought was that the back edge of the control ring (fuel dump ring) was worn, but i'm not so sure now. I'm fairly certain that the start of the injection control is ok (timing). Start and idle is good. It's the end of the injection that i think is the problem, but not sure what the cause is exactly. I just cannot believe the pump is toast, and, completley out of the blue.

i just don't want to go through all the hassle of pulling the pump, stripping it to find it's ok, then refitting it again. Nightmare job.

The symptoms i would best describe as running a diesel car with a clogged fuel filter.

Share this post


Link to post
Share on other sites

Do a leak off test on the injectors and change the fuel feed and return hoses to the pump to clear braided so you can see how much air you have in the fuel.

if you narrow the ticking noise down the pump then pretty safe to assume it isnt happy. Where was the swarf in the pump you found and how much? Drain the fuel filter and look for shiny metal flakes there as well.

Edited by Sausage

Share this post


Link to post
Share on other sites

I have two sets of injectors, both tested and are good.

i have run the engine from a clean fanta bottle with clear tubes full of diesel, to eliminate feed problems. No change...

The metal dust on the magnet, was just that- like a very fine black powder. No silver swarf anywhere. Quite a bit though, but not as much as you'd find on a gearbox magnet.

i'm beginning to think it might have something to do with timing. But from my basic understanding of how the system works, the no.3 injector monitors the start and end of the fuel pulse, by watching when and how long the needle lifts for, relative to crank timing. If the fuel timing were out, i'd have thought that the ecu would spot it straight away and flag a fault. I know there's a way to set the injection pump timing position by turning off the injection advance solenoid thing (which sets the moment at which injection starts), but i haven't cracked how to do it yet... the start of injection timing numbers i'm getting back from VCDS look ok. But maybe evrything in the system is being tricked somewhere which is why it might not be being picked up.

the thing that is really puzzling me, is that had i not tried the high rpm idle, with the injection quantity adjustment completely wrong (where the ecu couldn't run an idle map), i would have thought that the pump is 100% dead cert knackered. When the QA is completely out of adjustment, the car runs like it is on rocket fuel. Way more power than stock (and black smoke too). With my basic understanding of diesel pumps, that tells me that fuel is being injected- not just a little, but shed loads of it. The pump 'appears' to be working fine, or at least that's my theory.

when i pulled and retested the replacement set of good injectors, all nozzles were clear- no crud blocking anything,  telling me the fuel feed to the injectors is clean. The pump doesn't appear to be disintegrating......

i also might try running without N109 shutoff valve. Scary. No way to switch off if the engine goes open loop....

Share this post


Link to post
Share on other sites

buy a cheap spare sdi.....

Share this post


Link to post
Share on other sites

Yeah, i'm tempted rich, but it needs to be the £30 a year flavour.

i've been looking for an SDI Arosa. Did you know that there are only 43 left on the road... at least i'd get a glove box.

Share this post


Link to post
Share on other sites

ugly dashboard though.

buy it for spare parts, add up what a pump will cost and the price of a gearbox when that goes etc...

Share this post


Link to post
Share on other sites

Tempted.... :)

But, I am enjoying going through every single part of the internal design on a Bosch VE diesel pump.

I think I get how the N108 coil works now. It's just another servo system... Most explanations are rubbish. All it does is in effect is rotate the whole pump dynamically during running, according to a look up table (map) inside the ECU. The coil pulls against a spring. More ECU volts means more advance. The back end of the coil (the bit that slides in and out of th coil) is attached to a ring that has the cam rollers on. So in effect, by moving the rollers axially around the main shaft either way a bit, you can advance or retard the injection start point around a standard mid position.

Was out earlier and came back with a hot engine, so thought, it's time to do more tests... see pic below.

I've been reading how physical pump position timing isn't that critical, but it is interesting how you can use the measured and calculated numbers reported by VCDS, to work out how far out the three mounting bolts are. By looking at the requested and actual timing numbers; and PWM power, you can instantly see how far from ideal the pump has been physically bolted to the block. For example, if the PWM is really high (as in- the ECU is throwing lots of volts at the coil to get the valve to move, to give it as much advance as possible), it means that the position of the physical pump is set too retarded- the ecu has to work harder to get the right amount of advance. In the opposite case, if the PWM is zero or really low, it means that the physical pump position is too advanced.

So the readings below tell me that the measured injection timing has been adjusted by the ecu by powering the coil to 60% of full scale. Which is good enough I'd say. At mid RPMs with a hot engine, warm fuel and warm air, the N108 coil is operating just above mid point. It just ocurred to me- it has always bugged me why- when you kill the ignition at high revs then power on again, you get an awful din for a moment while the injectors are working mechanically, purely based on the physical position of the pump; Naturally, until the ECU wakes up, the timing will be out..... Aha!

 

timing.jpg

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now


×