300Tdi VGT project

[TheRecklessEngineer]

^^^ What Lara said - if the vanes are not moving 'till 1.1bar then no matter what spring you put on it won't make any more boost earlier. This does seem a little poor for a VGT - I have a standard Garrett stage 2 (whatever that is!) on mine and get full boost at about 1800....

[Turbocharger]

Cheers guys - it's good to get a perspective from someone who's done it before.

In the current setup with the diaphragm on its 'zero pressure' stop, the linkage is adjusted so the vanes are about halfway "open". It can make pressure much earlier if I set the vanes to fully "closed" (small turbo) - you can hear the compressor spinning up on just a high idle with closed vanes in one of my videos. However, in this position it only generates a very small pressure before going into surge as soon as I prod the throttle. This is why I've had to push the 'zero point' out nearly half way, so that it doesn't surge in that position.

From what you've said I'll try a more aggressive (ie more closed, smaller turbo) setting for when the diaphragm's at rest but I think it'll surge. I'm already a little concerned because the compressor's noisy, it's a very 'dirty' sound rather than a pure whine/whistle, but as it passes 1 bar and the vanes start to move it becomes considerably quieter and sounds happier - I think this is the onset of surging being quashed by moving the vanes out. You may be able to hear it in the dashboard video above as it passes 1 bar? The phone doesn't really do it full justice in reproducing the tone but it definately shows the reducing volume.

I'm know a weaker spring will move earlier and I'm hoping this will lift it out of the surge region gradually as pressure builds.

[Turbocharger]

I will add - the standard turbo would be up and boosting strongly by 2000rpm. The object of this project is to give more boost low down for driveability and eventually to fit the 1.2 tbox I rebuilt 12months ago, to give better fuel economy, lower noise and longer legs on the motorway. At the moment, it's worse than standard altho the idea's not dead yet.

I've certainly seen the potential for strong power benefits from lifting the boost level - if I can't see boost at lower revs from the VGT then the project's failed and I'll go back to a standard turbo running more boost and pocket the difference in insurance premium!

[Lara]

Hi John,

Difficult to tell by the vid but I personally doubt that you are hearing "Serge" Serge usually sounds like a slush pump and is not constant but a churning noise, it will almost certainly only be heard coming from the compressor side. Just think you have an old turbo, are hearing other sounds and not to worry.

Much more worrying if you hear surge type noises at low engine speeds and high throttle / boost settings.

Then start to investigate.

Lara.

[TheRecklessEngineer]

Could be cavitation noise of some kind?

Did I mention control theory at all? Surging is an unstable system - too much gain, too stiff a spring!

P.S. I'm not a control engineer, just covered 6 modules in my degree and hated every one!

[Turbocharger]

I sat through the same lectures, probably understood even less and no, I've produced no Bode or Nyquist diagrams for the system but I think it'd have to be considerably faster to start to oscillate - for the response I need the whole thing will operate in quasi-steady state anyway.

Surging is an unstable effect but it's not driven by the control system on the turbo, regardless of the gain. Surging is where each blade effectively stalls in the airflow - there isn't enough airspeed to generate the lift required by the angle of attack. As the pressure difference over the blade is able to equalise around the leading edge, when it's suddenly able to start building a pressure difference again until it becomes too great, stalls and starts again, so the compressor effectively switches very quickly between these two operating points. This alternating load on the blade can snap lumps off and that's why it's best avoided. (I did the turbomachinery lectures too and understood rather more of them...)

In this case the 'airspeed' is mass flow, the lift is a pressure rise across the blade which induces more flow and the angle of attack is defined by the compressor geometry - I can't just 'dip the nose' like a plane would to get out of the stall but I can reduce the 'lift' required by backing off the throttle.

I know the diaphragm isn't oscillating wildly because 1) the boost is stable at speed, and 2) without the bonnet I could see the arm directly and it was steady

If the weather warms up above minus five I'll nip outside and wind the adjuster back to a more 'interesting' position and give it another run.

[TheRecklessEngineer]

Excellent point - I didn't know VGT turbos could suffer from this. I stand corrected

[Turbocharger]

Well, after a chat with my dear old Dad he was a little confused too - he develops diesel engines for a living (and in fact, found me the turbo). His thought was that the lack of boost is due to limited turbine energy, for which I need more massflow (ie higher revs) or more energy in the gas (more fuel). That made sense since the EGTs on motorway cruise are much lower than before - there's much more air going in but still the same amount of fuel.

SOOO I'll have a play with the fuelling and see if that can push it up to higher boosts at lower revs. I'm not aiming for a revvy engine, I want torque and lots of it!

[Lara]

Hi John,

Have had a good look at all your turbo pics etc throughout this thread and would almost guarantee you that your turbo is not surging!

Exhaust is easily big enough and compressor side is in my opinion perfect for a 1 bar application! you could even go larger in all probability!

I believe you are thinking too hard Don't worry, set it to closed at idle and Drive it! You will see

Lara.

[Turbocharger]

Well, I've done some more playing. This graph shows revs against boost:

The red line is how it was, my baseline setting.

After talking to my dear old Dad I wound the fuel up half a turn on the base fuel screw (didn't touch the injector pump diaphragm) - this gave the green line.

Then I wound the vanes closed by ~1.5mm, still with the fuel turned up. This gave the response shown as the blue line... but the area shown in the circle was definately showing surge, a fluttering sound as if it was making boost for half the time at about 2Hz, sounding a bit like a misfire might. Sorry Julian Any idea what the boost response from your Td5 might look like on these axes?

The blue line hits 1 bar at 1800rpm which is about what I got from the old turbo - progress, but not the quantum leap I was hoping for. It's worth mentioning, running the engine on the blue curve at high revs means it gets to sixty in about 13 secs, which is frenetic and exciting but it's surging hard and it's absolutely not what I was aiming for with this conversion. It does suggest what could be achieved by winding the boost and fuel up on a standard engine though, EGT allowing...

I'm pretty sure I need a more proportionate response from the diaphragm to move the vanes, and I should have a handful of springs turning up in time for the weekend, if Mr MOT doesn't condemn me to a weekend putting new brakes on the MG.

[TheRecklessEngineer]

13 seconds eh? Sounds like I need to do some fiddling with my turbo

I have to say the mechanics of stalling a turbine is beyond me, but my gut would point me in the same direction - making the diaphragm give a more proportionate response...

[Lara]

Morning John,

Firstly, if you really are surging at 2500 rpm then you need to change your turbo!

Compressor surge is caused by too low air flow and too high boost pressure, Too much reverse torque on your compressor shaft with not enough exhaust back pressure, thus causing your turbo to stall, boost drops, turbo re-accelerates and thus re-enters the cycle. "serge" At 2500 you shouldn't be near either!

Can't think for the life of me why you would be getting Turbine / vane dynamic problems at this speed? bit if you are then there is a severe Miss match somewhere!

What is your Turbo? and what was it previously fitted too?

My curve would be as follows.

1100 rpm = Boost starting

1200 rpm = 5-6psi

1300 rpm = 8-9psi

1400 rpm = 10-12psi

1500rpm = 13-15 Boost pressure climbing real fast now!

1600rpm = 17-19

1700rpm = 22-24psi

1800rpm = 26-27psi Full boost!

No audible serge at any point.

Hope this is of some help.

Lara

P.S.

What happens to the noises if you set the poppet valve to closed at idle?

[DaveSIIA]

... but the area shown in the circle was definately showing surge, a fluttering sound as if it was making boost for half the time at about 2Hz, sounding a bit like a misfire might.

Bit of an off the wall thought - is there a possibility that the vanes are being driven by the dynamic response of the linkage from the actuator? Any mechanical resonance of the linkage could cause the vanes to move and hence alter the turbo characteristic.

[Lara]

That sounds like a very possible scenario, all you need is a corresponding frequency and off you go!!!

Fingers crossed.

Lara.

Although not perfect, would this not be a good idea?

[Bush65]

Good thinking DaveSIIA.

Looking at this pic, it would be easy to dramatically reduce the mass of the linkage. Changing the mass will change the natural frequency.

If it changes the revs where the surge problem occurs, that would indicate what is happening.

It does seem that the mounting bracket for the actuator and the linkage could all be improved if the turbo is going to be permanent.

[Turbocharger]

The linkage will certainly be improved if it's a permanent fixture. At the moment the poor stiffness is a useful feature because it means I can bend it into the shape I need so that the adjusters work over the right range. It'll then form the template for something more permanent. In truth, it's pretty stiff so it's not wobbling all over the place.

I see your argument over harmonic vibration, but wouldn't it tend to resonate equally either side of the static position - eg giving a generally correct position to the vanes? Then it'd need quite a lot of energy to move the whole thing at 2Hz and I didn't see a big difference when the M10 threaded bar was 300mm long (I trimmed it in situ). In fact, with the bonnet off there wasn't an obvious vibration problem.

I'm pretty sure that what I'm hearing is surge, but I've got a handful of springs in the post today so I'll try to have a play next week. First though, after today's events I have a disagreement with my MG's MOT man over the phrase "in the examiner's opinion" so I have to do a two hour job whole weekend's work first.

Keep the ideas coming though - they, and the tangible improvement in power, are keeping my interest in the project.

[DaveSIIA]

I see your argument over harmonic vibration, but wouldn't it tend to resonate equally either side of the static position - eg giving a generally correct position to the vanes? Then it'd need quite a lot of energy to move the whole thing at 2Hz and I didn't see a big difference when the M10 threaded bar was 300mm long (I trimmed it in situ). In fact, with the bonnet off there wasn't an obvious vibration problem.

The linkage may resonate equally either side of the static position (in terms of displacement response), but that might not be the primary cause of the "surge" effect. Looking at the structure of the linkage; the out of line forces; boundary conditions presented by the diaphragm unit and the vane actuator, suggests a non-linear element in a feedback loop.

Your graph of boost pressure vs rpm shows that the "surge" occurs in the transition region from rising boost with rpm to constant boost. In this region, the diaphragm will be moving off the end stop, so the boundary conditions presented to the linkage will change quite markedly. In turn, this will alter the dynamic characteristic of the linkage, so modifying the feedback system.

The other matter to consider is what is providing the mechanical stimulus to the linkage. At 2400rpm the rotational frequency is 40Hz and the firing frequency is 80Hz so this would not appear to excite a 2Hz "surge". The firing pulse is an impulse, rather than a sinusoidal motion. An impulse excites a much wider range of frequencies than sinusoid, so it is more likely that the linkage will respond at its natural frequency.

I haven't done any analysis to assess the boundary conditions or the response of the linkage. However, my experience suggests that a direct, linear system would enhance control of the vanes. It might mean clocking the compressor housing and remounting the diaphragm capsule, but the effort would eliminate many of the unknowns in the control mechanism.

[Turbocharger]

Well thanks very much Dave. You've ruined a perfectly good discussion with ideas I've been warned about before: "analysis", "critical thinking" and generally being probably right.

I hadn't linked the surging and the diaphragm lifting off its end stop - a strong argument. I can't clock the compressor housing because the pipes would point the wrong way, and because the inaccessible bolts are seized and I'm not prepared to risk it. Thinking laterally tho, there's no reason I can't move the diaphragm a little lower and angle it to point directly at the vane lever, much like Lara's dimensioned and toleranced CAD drawing above Looking at it now, there's no good reason why it has to point back along the car at all.

[DaveSIIA]

Oops, sorry

[Turbocharger]

Well, I have a package of springs appeared - £1.50 each but minimum order of £10 so I bought a few:

I put in a linear spring, which should start to move the vanes out as soon as it starts to build boost, at a rate determined by the spring stiffness.

I took it for a run, of which there's a video of the latter half. It makes boost at low revs, but I can't put my foot down too far because it's making too much boost - the spring stiffness isn't suited to opening up the turbo so it's hard to stop it making nearly 2 bar of boost, all the time!

To compare what it'll do to Lara's numbers above, (yes, you guessed) here's another graph:

The green bar shows how much more usable rev range I've gained - I can't stress how much difference is made by each lower 100rpm that boost comes in.

This looks excellent but the video shows there's heavy surging (I think) and it's definately not happy. Oh, and it's not really driveable because there's plenty of air going in but I daren't put my foot down to add fuel, so it's actually quite slow. I've now got the confidence to go back to the original spring with the turbo more closed... or to use one of the plethora of other springs to further justify my investment. The lessons learned from ten mins driving it this evening has justified that £10 though, I understand more about how it works with every change.

Has anyone done much work on what is an acceptable level of boost for a 300Tdi? Will it spit bits out at 2 bar? I don't think I want to run as much boost as Lara, but 1 bar across a wider rev range is quite acceptable.

[Turbocharger]

Hmm. During last night's forum outage I managed to post the above, but the picture of all my springs has gone AWOL and the topic's still on the second page. Hopefully this post will bump it back to the front.

The omitted spring pic:

As it says above, anyone know how much boost a 300Tdi will take before exploding?

[Pete Attryde]

IIRC the factory Discovery 300TDi boost is 1.2bar.

Mine is running 1.3 with no problems so far . (desperately searches for a bit of wood).

Pete.

[DaveSIIA]

There might be some mileage in using a softer spring, with the linkage set so the vanes are closed at idle. Theory being that the pressure will build sooner with the vanes closed, but the vanes will open more readily and limit max boost sooner.

I think you are on the right track though, with a (initial) limit of 1.0 - 1.5 bar boost while you explore spring/pre-load effects. At least you have an EGT gauge to help keep an eye on what's happening.

Does the adjusting screw on the capsule move the diaphragm/actuator to set the static load (pre-load) on the spring? If so, you might have a bit of juggling to do when setting the boost limit and vane initial position.

A bit of a random thought: Do you have any means of applying a repeatable, steady pressure to the capsule for calibration purposes? If so, could you set the spring rate and pre-load so that the actuator rod is just lifting off the end stop at say 1.0 bar and achieves full travel of the vanes by say 1.5 bar (these pressure values are guestimates)? You would then have to set the actuator length so the vanes were in the correct position. This would bring a degree of methodology to the selection of spring and it's initial setting.

[Turbocharger]

Thanks Dave - I went through the calibration process you suggest with the original spring:

I repiped the diaphragm since there was a tiny air leak which was losing "not a lot" of volume but "oh plenty" of pressure. Once airtight, I pressurised it with my workshop airline:

and then measured the diaphragm's movement:

(See, top science in this game).

but I've not done it for the other two - they're on the list but the garage is rather full of another broken car at the moment so it'll have to wait. Your suggested setup (lifting off at 1.0 bar and fully open by 1.5 bar) is what the original spring achieves (it's a wastegate spring from a Td5) but it was surging as it lifted off the stop. I've tried a spring today which lifts off at 0.2 bar and is fully open by 1.1 bar, the put in a spring in which my maths says should lift off at 0.8 bar and be fully open by 1.2 bar - both allow loony boost (1.6, 1.8, 2.0 bar) and I daren't put my foot down to add fuel because it's already making plenty of boost at very low revs. There's something wrong with my maths and I need to go through the calibration process above to understand more about it before I play too much and damage the engine.

I'm so nearly there - I'm getting plenty of boost across the speed range, but not all with the same settings under the bonnet. I haven't yet got a setting which can build boost at low revs without massively overboosting at higher revs.

[David Sparkes]

I haven't yet got a setting which can build boost at low revs without massively overboosting at higher revs.

In post 110 you showed the effects of increasing the fuel delivery.

Are you still running that increased fuelling?

Whether you are or you aren't, it strikes me that reducing the fuel, even below standard, will help stop you going into the overboost situation. At the moment you are gaining performance which you daren't use, for fear of totalling the engine. Reducing the fuelling will probably enable you to use more of the efficiency gains from the VNT, while not losing performance.

If nothing else, bringing down the maximum boost by reducing the fuelling will enable you to get a more complete picture of the effect of your control spring changes across all the rev range.

Reducing the fuelling will also take you away from the area of the graph where you noticed surge (or whatever it is).

HTH