Exhaust Manifold.

[Mr Whippy]

All I can say is, take ALL advice on the interbobs with a very large helping of salt... The amount of times I've assumed "It won't be that because x on y website says it's good for z" and then to realise infact it's all a pile of shit and they were merely speculating from very limited knowledge and understanding...

IMHO don't get too hung up on worrying whether you're going to affect the drive-ability and keeping things within stock limits... Just because a Diesel is rapidly dropping it's efficiency past 3500rpm, doesn't mean to say it's not good to rev it up to there with injection quantities still high... Every engine has an efficiency band, pushing outside of that for PERFORMANCE I don't think is a bad thing...



That above is how graphs SHOULD be looking... None of this power peaking at 3500rpm - that's what gives Diesels their crap reputation for "dying off" and "you're better off changing up a gear"....

Dave just out of interest - are you happy to map in bigger nozzles if you have calibration/flow data for the nozzles? Because as I see things - HDis seem to be in a constant battle between RP vs Durations... Because the nozzles really are so small as stock, the RP is through the ROOF trying to get the durations down, which equals more heat into the Diesel, running the HP pump VERY hard to achieve the pressure required... Bigger nozzles would equal shorter durations, less pressure required to achieve the short duration, therefore less power consumed to pressurize the fuel, therefore less heat in the fuel? VP pumped TDis have been running 220bar injection for years and are happily making 250+hp on simply upgraded nozzles whilst still returning 50+MPG...

But yeah - monitoring is one the most important tools in engine tuning - get a laptop running ELM327 and log your railpressure, get airflow logged... Also Johnny as said on the phone - upgraded air mass meter, then you can monitor your air intake, exhaust gas temp, exhaust pressure... Then you have a GOOD idea of what's going on

Again, don't want to repeat myself, but I agree with your angle.

Mod and learn, ignore the 'rules' on the internet because many are written/perpetuated by people who are basically wrong

BUT, when you do do something, make sure you can attribute any success or failure to that specific thing... and then do the community a favour by sharing that knowledge so they can avoid stupid crappy mods.

Ie, the whole cam thing for HDi's... that cam was never a good idea, even before I knew it's supposed origins I told people to check it's impact independently of any other mod, but no one did. They just threw it in with everything else and the net result of their work was impossible to guess at. Useless for them and everyone else. In most cases it was just £200 spent for no benefit at their intended power output. £200 that would have been there to recon the injectors when they failed and made the project have to be aborted perhaps... I dunno.

You only need to see that Pete has had 205bhp from a VNT HDi on a simple manifold and cam to know that you can do well without touching either cam/manifold to at least 200bhp.




Totally agree on the diesel tuning too. Just go look at a tractor pulling competition. They use thermal inertia in the turbo to run them hot as hell for the small window they run in. They run waaayyy beyond stoich. Stupidly inefficient BUT they still make more power, so in the fuel goes!
Revs, well, to a certain point yes, rev to 5000rpm. If your turbo delivery means you have a power band that means revving to 5000rpm makes sense, then do it. That is a diesel engine fully optimised for peak power if that is your goal (I'd take the 1.8T petrol tuned in that car personally but hey)

Obviously everything depends on the customers needs and you need to use your knowledge to provide them with the best compromise to fit that.



Power peaking at 3500rpm is simply a function of torque though isn't it. If it made the same power at 4500rpm, or 5000rpm, it'd be no faster at all. It'd just cost you more money and be less efficient.
I agree it's not ideal, you make your gearing shorter in real terms which is the biggest cost.

Really the turbo is the limit, it's small and flows well to a point. If you could make enough flow to make 150bhp at 4000rpm you would, but you can't. So you make a curve with as much area under it so a customer can go as fast as possible.

Curve shaping is important but many customers like the punch of near 300lbft, so even though it makes the delivery odd, it's what they want. Personally I prefer the cars with less torque to artificially shape the curve more flat and peaky revvy feeling.




The HP vs RP thing is kinda an issue. People chase RP to compensate for long durations because of small nozzles, but that is only half the issue.

The way I see it.

I've seen 300lbft of fuelling on a standard pump/injectors at 2000rpm. Ample for most users needs.
I've seen about 175bhp at 4000rpm (on non-ideal mapping) with standard pump/injectors.

Having bigger nozzles won't remove the flow issue though. In the end to get power you need fuel. Throw it in a 1000000bar or 1bar, the issue is getting in the desired amount in a useful time.

So really the issue is pump flow and as above, it's flow probably runs out of juice around 180bhp at 4000rpm, or 300lbft at 2000rpm.

At a certain point adding more RP to get IQ up is just the same to the pump as opening the injector for the same time with bigger tips. The rail pressure drops, and if it drops too far then the pump can't flow enough and the injector isn't injecting the desired IQ in the opening time given.

Not sure if I've explained it really well but basically when the pump is at it's limit the injectors are irrelevant.

But then we need to consider what you mentioned. Efficiency. If we make efficiency better we can make more power per unit fuel. So we can boost power other ways for the same injection. Ie, less back pressure, manifold porting etc. If they make sense to do that is
Of course you know all this already. But it makes sense to re-iterate that there is no point moving up until we sensibly address issues at each step. Going right for a ported manifold or head ports when we have a crappy exhaust or little turbo, or a limited HP pump, are all just pointless.





I've done a few HDi's with bigger nozzles, but, sorry to say it, they all failed to work well.
In almost all cases a combo of too many mods all at once, or dodgy components, made them impossible to tune properly with. Ie, in one case I spent a few weeks of solid tweaks despite Pete telling me I was within the right ball park with my initial calibration map to start with. Issue, some washers on the injectors were the wrong ones when they were rebuilt by the owner.
In another case, the HP pump wasn't any good, it was uprated but buggered. The customer just sold up and moved back to stage 2/standard injectors as a recon upgraded pump was beyond their budget.

The few projects that succeed are highlighted by the fact they take a long time to achieve and cost a lot of money to get everything working 100%

Usually the customer/tuner work up in steps and get each thing perfect before moving on to the next.

I don't charge customers per visit or per map, I charge for the project usually, or stage it to make sense for the mods they want to do. I try encourage people to work sensibly and logically through their projects to make it easier for them and me!

But clearly it still goes wrong due to big ambitions and a lack of funds and patience.

You can immediately see why the only place that ever packaged mods for these cars (Derv Doctor) didn't go buying used bits off eBay, just like every other big tuner out there they buy new/recon bits and put them on knowing it'll work right away and if it doesn't they just take bits back for a new one.

Obviously it's hugely expensive, so doing it DIY can save a lot of money, BUT, you have to take all those risks of bits not working as intended and sometimes it can cost a lot of money to get some of these exotic parts working right...!





If I were to build a HDi tomorrow to make 200bhp, I'd buy a HDi.

First step full service, cambelt, fuel filters, clean lines, RP relief valve, get everything running as perfect as I can standard.
Stage 1, full fancy exhaust. Run it for 3 months, let the car settle in and see how it goes. Fix ANY issue if it shows up.
Stage 2, standard clutch. Run for 3 months, fix any problems that arise.
Clutch, then push torque as far as it'll go, run for 3 months, fix any issues that arise.
Upgraded pump, fit and run for a month to make sure everything is fine.
Injector tips. Fit and spend as long as is needed recalibrating the maps to suit them.

Hybrid turbo, something that on a remapped road car runs about 200bhp with a good exhaust etc... maybe a GT17 type front end on a back end optimised for about 4500rpm peak engine speed.

Fit and push fuellings iteratively in 10bhp increments every week or so until I hit around 200bhp at 4250rpm or so. Check/fix anything that is a problem along the way.

I bet I'd be about 6 months in the making, and I'd probably have spent about £2000 by the end too... if not £2500.

Eeek. But I know I'd reach that target, and reach it well, and it'd work reliably without smoke, start perfectly in winter, my mum could jump in and drive it etc etc...



Big power HDi's are possible. There is no reason they are not. But not much knowledge exists beyond 200bhp.

The way to get 200bhp+ is to get everything that is known to work well done first, do it all well and make sure it all runs perfectly. Then work up in steps and tick off every mod along the way as a mod having been worth doing.

Manifold porting, before/after dyno, maybe some pressure checks. Tick. It works, if not fix it/revert.

De-MAF, or double MAF (parallel) for more flow, check it all works ok, tick. It works, if not, fix it/revert.

And so on. No need for doing maths/sums, just good old before/after and sensible logging to see if you are wasting time/money or doing something useful


Just throwing a massive combo of unknown quantity parts/mods at an engine and expecting amazing results is just not gonna work




There is no real answer in there I don't think, just my perspective on things.

I'm all for modding basically, and all for 'test and see' vs theory and maths. Just check before/after otherwise you may just be pissing into the wind!


Dave