Our Peugeot Diesel - Part 5 - Final Tuning of the Injection System

Last story in this series we looked at how the Bosch VE injection pump on the Peugeot 405 SRDT works. We also raised the boost level by adjusting the wastegate but the on-road results were a bit disappointing – the performance, while improved, didn’t seem to match the boost increase. However, all that changed in dramatic fashion when we started to get into the fuel pump adjustments!

Adjustments

When you start talking fuelling on a Bosch VE mechanical diesel injection pump, most people talk about just two adjustments.

The mixture adjustment screw (arrowed) allows addition or subtraction of fuel across the whole full-power rev range.

The preset of the boost enrichment diaphragm (the screw on top of canister) sets the off-boost fuel addition. (Note: this is correct, but it is NOT what most people believe!)

However, in performance applications, the most important adjustments are under the boost compensator cap.

Remove the screws (I replaced them with easily undone Allen bolts) and you’ll find a diaphragm.

The diaphragm can be rotated, so also rotating the control rod that adjusts the addition of on-boost fuel. And, because this control rod is eccentric, rotating the rod allows the amount of extra fuel added on boost to be altered in magnitude.

The other adjustment is a ratchet cog that allows the preload on the diaphragm spring to be altered. This in turn changes when the boost fuel enrichment starts to occur. This cog can be accessed by lifting one side of the diaphragm and then carefully using a small screwdriver to rotate the cog.

Considering that a well set-up diesel turbo is on boost most of the time it is working hard, these on-boost fuelling adjustments are extremely important.

On-Road Results

When I found the power that these adjustments had over the fuelling, I was as happy as a pig in sh.... Well, I was very happy.

I wound up the boost enrichment and found that with the wastegate adjusted as described last week, I could get no less than 12 psi boost in 4^th gear at just 2000 rpm. (That’s nearly double standard!) The improved bottom-end torque had to be felt to be believed. But wind back the fuel and that could easily drop to 4 psi – all without changing the turbo wastegate setting.

At the top end, adding sufficient fuel could generate something like 26 psi boost – or, with less fuel, that could be 18 psi. Again, all without changing the wastegate...

So you can imagine what I was by now doing – tipping-in as much on-boost fuel as possible while at the same time leaving off-boost mixtures suitably lean.

But there’s a problem with this approach – a big one. Too much fuel results in a diesel belching black smoke. I was using the rear vision mirror to check for smoke outputs and while I could see some smoke, it didn’t look excessive. In fact, I drove with these high-fuel settings for a day or two, even laying down a 0-100 km/h time of 11.5 seconds. (Compare that with the standard 15.3 seconds!)

However, one day I thought it might be good to drive along behind the Peugeot in another car, watching the tail-pipe smoke output.

And when I did this I was appalled.

Simply put, there was far too much smoke everywhere – at light loads, at heavy loads, on gear-changes... the lot. More than any other limitation, I found that getting good performance with low smoke outputs was the balancing act.

Many of the adjustments on the pump are lock-wired or are fitted with tamper-proof fittings. For example, this crimped-on collar was fitted to the main full load mixture adjustment. Removing the collar was easiest done by unscrewing the screw from the fuel pump unit, measuring the position from the end of the screw to the nut (so that the original setting could be easily returned to), screwing the nut out of the way then carefully grinding the collar sufficiently that it could be pried off. The nut was then returned to its original measured position, the screw carefully cleaned and then screwed back into place.

Tuning

After many hundreds of kilometres of road testing and tuning (in fact, nearer 1000 kilometres), I found the following tuning procedure best:

1) Disconnect the boost compensator and plug the manifold hose so boost cannot be lost.

2) Adjust the main mixture screw (use only very small movements) until a trace of smoke is visible under full throttle, and excessive smoke output does not occur at any point in a drive cycle that includes hills and flat roads. (Hardest in all the tuning is to achieve low smoke outputs coming on and off boost, eg in second or third gears when climbing undulating suburban roads, and when short-changing from first to second.)

3) When the main mixture adjustment is correct, reconnect the boost compensator.

4) Remove the boost compensator top and mark the original position of the diaphragm (eg with a magic marker texta). It is rotated clockwise 90 degrees for max fuel. Test drive the car, changing in steps the orientation of the diaphragm from 0-90 degrees to find mixtures that produce little full throttle smoke but cause best (highest) boost results. Boost should be measured at 2000 rpm full throttle 4^th gear, and also at maximum revs and throttle, in second or third gears. This effectively shows the boost levels at each end of the engine’s on-boost rev range. (Remember, this is the wastegate closed sufficiently that boost is primarily determined by fuel addition.)

5) If the fuel appears to come on too early or too late, rotate the spring ratchet cog and assess results.

Expressed in this way it all looks pretty straightforward – and by following the above test sequence and being patient and careful, it is pretty straightforward.

I found that by far the best way of assessing smoke levels was to have a following car. If both the drivers of the car being tuned and of the following car have two-way radios, continuous communication can be maintained. In this way, a real-time commentary on smoke outputs can be made by the following driver – this quickly shows when smoke is being produced – eg, on the transition to boost, when revving-out under full boost, on gear-changes, and so on. Without this immediate feedback, it is very easy when tuning to chase your tail.

Please believe me when I say that as the driver, you simply cannot accurately assess the smoke outputs by using the rear vision mirror!

One diesel specialist told me that for AUD$800, they could set the pump up on a test bench so that it would immediately give the desired on-road results. However, I find this very hard to believe, given that the turbo boost pressure that’s developed depends so much on the fuel being injected, which in turn influences the power being developed, which in turn influences the energy available to drive the turbo, which in turn influences the engine airflow, which in turn influences the amount of smoke produced! Furthermore, the actual intake and exhaust flows will depend on the intake and exhaust modifications that have been made. If the wastegate is positioned so that fuel is the limiting factor to the boost being developed, the goal becomes: achieving across the rev range maximum measured boost with minimum smoke. In addition to being carried out on the road, this tuning could also be achieved on the dyno. However, real world variability in driving will show up conditions for smoke production that a dyno will never show. For example, in my test drive cycle, there was a T-intersection that required a hill-start followed immediately by a 1st – 2nd gearchange. Getting low smoke production in just this one condition was quite challenging. So I think although it’s time-consuming and fiddly, on-road tuning while in communication with a following driver is likely to give the best real-world results.

Tricking the System

After extensive tuning I achieved these results:

• 7 psi of boost at 2000 rpm, full throttle in 4^th gear (standard boost in these conditions was about 4.5 psi)

• 19 psi of boost at full throttle, maximum revs in 3^rd gear (standard was about 13 psi)

• no detectable smoke when idling and at light throttle

• a trace of smoke under normal driving

• what I deem acceptable smoke output under full acceleration*

*So what is “acceptable smoke output” under full acceleration? Where I live there is no annual test for smoke emissions and you’d be very unlikely to be pulled over and have the car defected for a smoky exhaust. I have therefore set the full acceleration smoke output at a level that’s not noticeable in daylight to the driver (but can be seen in the headlights of a following car at night) and can be seen - but probably without comment - by a following driver. Remember, this output is only when the accelerator is mashed to the floor.

However, after this tuning was finished, a problem remained. The issue was the puff of smoke that occurred on gearchanges. If the boost enrichment was leaned out, this could be eradicated – but then mixtures were too lean under high load to develop the desired boost levels.

The way the gear-change puff of smoke was eradicated (or, depending on the conditions, vastly reduced in size), was to change the pneumatic signal reaching the boost compensator. By placing a needle valve in the line from the manifold to the boost compensator, and then after the needle valve placing a T-piece that connected to a closed reservoir, the rate of pressure increase reaching the boost compensator could be adjusted by altering the needle valve opening.

The system works like this: before the boost pressure fuelling compensator starts to see boost, the air has to flow through the restrictor (comprising the needle valve) and then fill the reservoir. This slows the action of the boost fuel compensator on transients, while still allowing it to add fuel normally when the boost is sustained.

The reservoir that was used was an aluminium EF Falcon fuel filter fitted with a threaded plug at one end and a barbed fitting at the other. This reservoir is good because it’s readily available, copes with the pressure and looks good.

I already had the needle valve and some of the fittings, so assessing the cost of this mod is hard but it should be able to be replicated for perhaps AUD$70. Needle valves and fittings are available from hydraulics and pneumatics suppliers – check the yellow pages.

Results

The results are interesting. Firstly it must be stressed that these final results are way slower than I achieved in testing that had lots of smoke output. In other words, if you’re unconcerned about smoke outputs, far better performance results can be achieved.

The 80 – 100 km/h time in 4^th gear is now 5.5 seconds – 0.3 of a second better than after I first altered the wastegate setting and increased the overall fuelling. (And not far off twice as quick as the standard car!). Top-end performance has now improved – the 0-100 km/h time is now 13.4 seconds, better than the 14.0 seconds recorded when the fuel was added across the whole range and clearly much better than the standard 15.3 seconds. (However, obviously the Pug is still quite slow.)

But it’s the driveability which is really the outstanding improvement. In normal urban driving you’re now able to use a full gear higher than with the car standard. In standard guise, 5^th gear was for 80 km/h and above – now, you can be happily trundling along in 5^th at 65 km/h, or about 1500 rpm. Fourth gear is now great down to 50 km/h.

These driveability figures don’t sound very inspiring but the extra torque that allows the car to pull fifth at 1500 rpm now means that open-road hills that previously required a change down to 4^th gear are now taken in fifth; hills that used to require 3^rd gear can now be driven up in 4^th. It’s now rare to do what was very common with the car standard – to change up to 4^th or 5^th gears and then find the engine cannot pull the selected gear, and be forced to down-change. These changes are even more pronounced over standard when the air con is on.

So while outright full-throttle performance is now improved and the engine revs far more freely to the redline, it’s the part-throttle and low rpm performance changes that are most improved on the road.

Intake Air Temperature The Bosch VE system does not use any intake air temperature compensation - therefore, fuelling does not alter with air density. One interesting result of this is that when the under-bonnet intercooler has heat-soaked, the car will smoke far more than when the intercooler has come down in temperature. This is because as its temperature rises, the mass of air being breathed decreases – but the fuel addition remains the same. So when tuning the mixtures, you need to always make sure that the intercooler is at normal temperature before assessing smoke output. Also, a car tuned so that smoke is only just acceptable in normal use will have unacceptable smoke output for a minute or two when started hot. Finally, in normal driving, higher ambient temps and intercooler load (eg after climbing a long hill on a hot day) will result in more smoke. So when setting the fuel addition and assessing the smoke outputs, keep in mind the intake air temp.

On the Dyno

The last dyno run we recorded, done with the intake and exhaust modifications, showed a peak power of 61.5kW. The graph also showed that over the 56.6kW standard figure, power was most improved from 3000 – 4500 rpm. However, as noted at the time, the bottom-end improvement (eg around 2000 rpm), while quite clear on the road, was not seen on the dyno.

This graph shows the improvement achieved with boost and more fuel. The lower red line is the ‘intake and exhaust’ power figure and the upper green line the ‘boost and fuel’ results. As can be seen, the power output is lifted, especially from 3000 rpm to 4250 rpm.

(And what’s the uppermost red line? We’ll come back top that in a minute!)

This graph compares dead standard with the current power output. Peak power has risen by 14 per cent but what is more interesting is the gain at lower revs. At 3500 rpm, power is up by 16 per cent and at 3250 rpm, up by 20 per cent. Another way of stating this is that the dyno shows the engine has a 20 per cent torque increase at 3250 rpm. So while the peak power increase is modest, the area under the curve shows that average power through the rev range is well up. Additionally, the on-road results indicate a substantial gain down to at least 1500 rpm, something the dyno graph doesn’t show.

Now what about that red line on the graph above? On the dyno the underbonnet intercooler was getting very hot. So we lifted the bonnet (that in turn removes the ducted air going to it) and then during the dyno run, sprayed the intercooler with water from a pressurised garden sprayer.

As can be seen, there was an immediate 2.5 per cent gain in peak power – but a really radical 15 per cent gain in power at 2500 rpm! This strongly implies that the intercooler got hot as the dyno run progressed, and that starting at a cooler temp (caused by the water spray) helped the power output considerably. In turn, this suggests that better intercooling could make a substantial difference to power.

Looking also at the improvement in lower rpm compared with upper rpm, the extra boost seems to make a big difference down low but not up high. Amongst other factors (like turbo sizing), this could also point to an inefficient intercooler.

Fuel Economy

At the time of writing, fuel economy of the modified Peugeot looks to be about 4 per cent better than when the car was standard. Obviously, the free-er flowing exhaust and intake, and the ability to pull taller gearing, more than outweigh the greater consumption at high loads.

Conclusion

This article concludes the first series of mods to the Peugeot. To recapitulate:

• The exhaust has been upgraded with an ex-Commodore V6 rear muffler

• The intake system now uses a Commodore airbox, EL Falcon intake bits and pieces, and a new convoluted tube intake to the airbox

• The wastegate adjustment has been screwed fully clockwise, with the fuel addition now determining the realised boost pressure

• The diesel mechanical injection fuel pump has been adjusted to give a higher rev limit, a little more fuel throughout the full-load range, and much more fuel when on high boost

• A pneumatic tweak has been applied to the boost part of the fuel pump, reducing smoke emissions on gearchanges.

	0-100 km/h	80-100 km/h (4th gear)	Fuel Economy (JE normal use)
Standard	15.3 sec	10.7 sec	6.9 litres/100km
Modified	13.4 sec	5.5 sec	6.6 litres/100km

The total cost of these modifications was well under AUD$200, including the fuel used in the road tuning. The car is completely transformed, especially in driveability. Where outright power is needed – eg towing a trailer uphill or open-road overtaking – the Peugeot is clearly no ball of fire. However, in normal urban and country road use, the car is more than acceptable. And, even in the context of current cars, the fuel economy is excellent.

Dyno runs courtesy ChipTorque.

Share this Article: 

Digg this story

Seed Newsvine

Post to Reddit

Save on del.icio.us

Email