Water injection in TD5 - anyone done it?

[Cret]

Thanks for the comments David.

That's a useful bit of info in terms of the flow of the exhaust gas influencing the best theoretical location of the probe, although I know from experience that this doesn't always translate to somewhere that's practical to fit something unfortunately.

What you've said is perfectly sensible in terms of engine variations etc influencing readings and so on.

What I usually look for in any warning instruments is a variation from the norm, so I'd be testing it all on a bog standard engine run (ie chip removed & no WI in use) then compare afterwards to analyse the results.

The WI kit I have is an Aquamist one - it's a great quality piece of kit although it's not actually installed & in use yet.

What would be interesting to me would be photos of where different people have situated the probe for EGT gauges they have. Normally with these sort of things there ends up being a favoured spot.

Is the manifold easy to remove on the TD5? If it is, then I could potentially look at the fit over the studs as you mentioned, and potentially add a little bit of extra clearance if it looks like an issue.

Cheers

Jim

[nicks90]

jsut had a couple of thoughts going round my head about this and have a question or two...

doesnt water injection REDUCE the pressure depending on when you inject the water mist?

how to explain?????

say the air temp is 20C at the inlet. it gets compressed by the turbo and temp rises to say 80C and 18psi at the actuator on the end of the compressor housing.

As the 18psi air goes through the intercooler is gets cooled to say 50C. Wont the pressure drop as it travels through the intercooler? As the hot air cools it contracts and becomes denser. But as the boost is set at the end of the compressor wheel, the turbo wont account for this?

After the intercooler you inject water spray which reduces the temp to say 40C.

again, you have reduced the temp further - so wont the air density rise - but pressure drop again?

If my logic is correct you end up with initial charge of 18psi boost at 80c dropping to 40c and maybe 14psi at the inlet valves? The physical amount of oxygen entering the cylinder wont change, as the physical mass of oxygen is the same between 18psi@80c and 14psi@40c...

I presume this can be countered by putting the boost actuator on the inlet manifold (like you can do on tdi engines so you get the full boost pressure at the inlet instead of at the turbo),

But i'm just wondering if my logic is correct or complete misunderstanding of bernoulli's law.

[Cret]

It's not something I've considered, and I can understand the argument you're making so it's an interesting question.

I'm inclined to agree with your logic, but what we all do know is that intercoolers make a huge difference to performance by reducing the temp of the inlet charge. That's the bottom line here so we know that your theory notwithstanding, they do work.

I don't think it has a detrimental effect even though on initial impressions it might seem like it.

As I understand it actual pressure is not of overall importance in a turbo system. This may sound odd but allow me to elaborate.

The overall idea with a turbo is to force in a greater volume of air in. That's the objective and the result.

The means of doing this is to pressurise it, but for instance if you run a ginormous turbo (lag/spool issues aside) you get the same volume of air into the engine as with a small turbo straining its nuts off, but at a much lower pressure. It's the actual airflow/cfm that counts, not the pressure per se.

This is for instance why on my legacy I can upgrade the turbo and run the same boost pressure before/after but get an increase in airflow & thus can increase the fuelling to match via my mapping.

The best analogy I've heard for this is to imagine a hose filling a bucket with water. You can use a small hose with high pressure water and fill it up say in 1 minute, or you can use a larger bore hose with a lower pressure output of water and fill it in the same time (ie a big turbo), so the pressure is not really the issue.

So yes, I think you're right that the pressure may reduce as you've said, but this does not matter because the volume of air is still there, and that's what dictates the fuel burn & ability to add more fuel etc.

As for the amount of oxygen, the addition of water vapour enhances this, not so much the actual cooling. Or something. I know the water vapour does this, but my head is actually struggling a bit with whether the actual cooling of the air on its own provides more. All things being equal I guess not (I'm not a chemist nor a physician though!), but I can't get a lid on whether this matters because of the constant flow of air coming in.

Hope that makes some sense, but even if it doesn't as I say the bottom line is that cooling the inlet charge does of course help as we know, otherwise our cars wouldn't have intercoolers.

What do you think?

[Lara]

Hi Cret,

Simple answer,

Go for it!

Try it and experiment, but probably best benefit will be in the 1.3 Bar + range as your inlet temp will be substantial.

A lot of explanations why things do and do not work have been mulled over on this thread but.

The reason any theory does not stand up to practice is because it is wrong!

We all know that an efficient inter-cooler adds power even at the cost of lower boost, when theorizing, often we forget that nothing ever stays the same, (we rarely change only one thing without affecting something else) i.e. If we lower the inlet pressure anywhere in the system, The turbo will speed up and more FLOW will occur as the higher pressure air will flow towards the lower pressure faster as there is now a higher pressure differential than before* (remember our turbo wastegate is only pressure sensitive) *(unless the turbo is at it's limit or something is restricting the flow) This is just one reason for some theories to be in error. This knock on effect will go on "add-infinitum" and it not only makes it very hard to calculate and theorize it makes it extremely hard to explain

It is very hard to account for everything and there are so many variables, and so many estimates that make things extra difficult, i.e inlet temps, how do we know what this is going to be outside the lab? we have to estimate and test! we can calculate our "Y" values etc but as these depend on initial ambient temp, and as this changes every second in England, what do we do?

Go and test! Have fun! and learn in real life! more useful, more educational and a hell of a lot more interesting!!

and while you are at it, mix a bit of Methanol in with the water and test that!!

Lara

P.S.

This post is not in any way meant as a criticism of anyone else's posts!

[Cret]

Cheers Lara

Don't worry - go and try it was always the intention and still is, the thread purpose was primarily to get comments from people who have already tried this modification so that I can compare notes, see what findings they've had, and get a rough idea of the sort of setup that might work optimally.

I welcome the discussion from people who just want to find out more though of course (as you can probably see from long winded replies I make!).

I think I may have found the issue with the dastek I just got as well so hopefully that will give me extra potential for good results from the water.

Yes I plan to try methanol as well, although initially I want to see what water achieves without it.

Cheers for the comment though - you've basically explained what it was that was going on in my head in terms of resolving the oxygen quantity question etc & why I was struggling a little bit (plus it's been a brain frying week again at work and I was also concentrating on not scalding myself with the hot pie I was busy scoffing).

[66gaza]

Injecting windscreen washer fluid (water & methanol) is also known as chemical intercooling.

Gaza

[Cret]

Quite correct Gaza. Think I made a comment about this earlier on as the washer tank is the typical source of choice for the water tank, both because of convenience of already being there, and because washer fluid is (ish) 45% methanol. Handy indeed!

Little update - my mate is 'booked' to come over in the next few days to weld bits on my EGR replacement pipe as injector bosses etc. I can weld - sort of - but this needs to be neat.

Once that's done & they're drilled & tapped I can get it all in.

Other than that, I got the dastek working correctly today. Found the boost signal wire was a bit knackered so I sorted it out and bloody hell! The disco feels like it's transformed!! Lovely throttle response, goes a shedload better when you put your foot down, and not a whiff of smoke that I can see so far. Love it.

This should hopefully give the water kit a real chance to shine since it's a sort of intercooler equivalent and an intercooler is meant to give most benefit when combined with an increased tune.

Even if non of this makes much difference I'm dead chuffed with the performance and 'nice factor' the dastek has provided.

[drkwack]

Hi, been reading this with interest, now this may seem an odd thought, but do the top drag teams in the USA use any thing like this as they have fantastic sums of money to throw at things, seems like it may be an idea to contact them and see what comes up.ON the other hand the JCB land speed team for diesels might have a thought on the subject.I know an old mate did this on his Harley took some setting up and when it started properly he got an amazing cloud of steam and carp out of the exhaust, my other half commented " all that money and he's gone back to steam power", yep we had to do a runner until he saw funny side , best of luck with it.

[David Sparkes]

You might be interested in this article http://autospeed.com/cms/A_109423/article....?popularArticle

It's essentially about the different work rates required of a diesel v petrol intercooler installation.

However, the points it makes about the need (in a diesel installation) for constant dissipation of heat, and the benefits of avoiding heat soak, makes interesting reading inconjunction with enhancing the cooling of the intake charge.

From my own experience, if you have fans in front of the intercooler, running them at half speed on a hot day does help to inhibit heat build up in the intercooler, mainly because it brings the temperature down when the intercooler isn't being asked to work hard, but has little airflow, such as in slow moving traffic. Thus as the work rate increases the intercooler is starting from a lower temperature and can heat soak more energy from the compressed air.

The slow fan speeds (by putting the two fans in series) means you aren't asking too much from the alternator, and can thus run them 'all the time'.

The same magazine has done several articles testing the changes caused by spraying finely atomised water on the outside of the intercooler, tying the spray control into the throttle use (and it's not as simple as it first appears).

They have also recently done 2 articles on a new 5 channel data logger. This feeds into the USB port of a PC. There are limitations, but for only 87AUD (~42GPB) it seems good value for money, even when you add postage to that cost. Customs and Excise charges could kill it stone dead - that seems to be their task in life.

With a nod to the comments 'don't know how it works, but it seems to', the article makes the point that however you do it, if the engine breaths more effectively, more air will pass through the MAF. Monitoring and recording this (as a CSV file) really gives you before and after comparisons.

To me, monitoring the MAF seems like a good constant reference point, whether you are taking baseline readings, evaluating the Dastek, injecting water in the intake stream, changing the intercooler, or cooling the intercooler.

References to Parts 1 and 2 respectively.

http://www.autospeed.com/A_111416/hDg34uLt...ms/article.html

http://autospeed.com/cms/A_111417/article.html

Good Luck

[Cret]

Thanks for the reply again David

Going to have a read through those articles and see what's what, but at a glance they look quite interesting.

I've made a little progress last night anyway.

I now have two bosses welded onto my egr replacement pipe, one for the injector, and one for the pressure switch.

Blurry photo attached.

They need to be drilled & tapped next, and they're 'funny' size threads (ie not ones in my cheap & nasty tap & die set).

Luckily, aquamist supply a tap for the injector with their kit and I still have that, for some odd reason they don't appear to provide one for the pressure switch though. No idea why. The one I need is 0.9mm thread but am unsure about the width/size required as it measure a bit different to normal 'M' bolt sizes.

I might hand it over to a friend of a friend who is an engineer, and see if he can do it for me. It'll be a much better job than me doing it even if I get the right tap for the pressure switch.

*edit* Having cut & welded on the two bosses - I've just found out that Aquamist supply a remote mounting kit for the pressure switch with an adaptor that screws on to enable use of a barb tail fitting and small hose supply for the pressure feed. This negates the requirement of having this drilled & tapped so I can sort of get on with the installation if I do the other thread myself. All for a measly £6 too!

Think I'll be ordering one of those today then.....

[Aragorn]

A little pointer on the above comments about the turbo not seeing the lowered pressure, first of all, it will, as if you shrink the air further up the pipe, the pressure drops everywhere, so the turbo would then speed up and bring the pressure back up to the desired level.

Secondly you can always move the boost sense line, from the compressor housing to the inlet manifold, which will ensure you have the exactly right amount at the inlet.

[Cret]

Cheers - I've been planning on moving the boost takeoff to the inlet manifold side for that reason as that's where you get the usuable boost figures from.

It's on the turbo side at the moment for ease of fitment (the boost gauge takeoff I mean), but just poking my head out of the garage for a moment it doesn't look like there's a suitable takeoff point on the manifold area so I suppose that's more tapping/welding required to add a fitting.

[Cret]

Right! Injector side of things is sorted out now so I just need to get the bits to plumb in the pressure switch, then mount the kit.

The tricky stuff is (I think) done now!

[Cret]

Just read that autospeed intercooling article and it was a nice read. I've not really considered previously that the diesel has different requirements in the way it describes but it all is totally understandable.

If anything, I would expect that this increased desire for efficient cooling on the charge should mean that the water inection may given better results than I was originally expecting, especially as the car is now chipped & under increased pressure & therefore increased heat over standard.

Having read that, I like the idea of fans. I actually found today that I still have the pump, solenoid, and injector from my original cheap nasty water injection kit from several years ago, and these could be put to advantage by being used as a very fine misting spray on the front of the intercooler, like subaru use on certain STi Imprezas.

Was interesting to see mention of the very same chargecooler that I used to have on my Legacy too! The Liberty RS is the same car as the Legacy RS which I have, just the Australians called it Liberty in stead of Legacy because of some ww2 issue, and it is a lower tune, but the chargecooling system is what was originally on my car. I have upgraded this to a front mount which works better and does not have the problems of pumps failing etc. The front mount also causes some entertaining noises on closed throttle (petrol engines only obviously) where compressed air goes back over the compressor wheel, although it should be removed by a recirc dump valve really for turbo health.

I like the idea of the datalogger too. I do actually have a datalogger for the subaru's apexi powerFC ecu that I run & map myself. I doubt I could use it on the disco though. It's an FC Datalogit unit that I have: http://www.fc-datalogit.co.nz/

The plan of using MAF data is an interesting one indeed.

[Lara]

A little pointer on the above comments about the turbo not seeing the lowered pressure, first of all, it will, as if you shrink the air further up the pipe, the pressure drops everywhere, so the turbo would then speed up and bring the pressure back up to the desired level.

Secondly you can always move the boost sense line, from the compressor housing to the inlet manifold, which will ensure you have the exactly right amount at the inlet.

Hi Aragorn,

Not quite right as we are not talking about a sealed pressure container, we are actually talking about one with an almighty hole at one end and several restrictions in between,

As you measure the positive pressure all along the inlet system you do see quite a few differences, some small, some large, If you have a very small plenum chamber you can even measure quite heavy pulsing of this pressure as the valves open etc, especially with V twin engines because of their un-equal firing strokes! But you will see the "flow" moving toward that low pressure area to try to compensate! The turbo will probably eventually speed up enough to compensate but if you are on the efficiency limit of the turbo, it may not, or at least it may be very slow to do so.

Lara,

P.S.

Racing formulas running petrol / diesel etc usually prohibit the use of power boosters, and or alcohol etc injected as a fuel, although Drag cars do have classes for almost all fuels. although you would get little benefit with water injection on a Methanol burner due to the massive latent heat of vaporisation of methanol causing the inlet temp to be extremely low, probably too low to vaporise the water well enough.

[Cret]

I'm staying out of this one about the turbo flow!

[Lara]

It's not an argument, just a different way of explaining the same meaning really

P.S.

The Aquamist system is a good aftermarket kit, can be improved on but is a very good start.

Lara

[Cret]

I like the idea of upgrading to the next system up where you can map the water flow to a parameter from the engine to have a varying amount.

Can't afford to do that though, and I've yet to see if it works well yet.

Think I've resolved a minor issue tonight anyway. The remote mount kit I need for the pressure switch alters the switch from using a thread to a 4mm hose barb tail connector.

That's fine as I have a big coil of 4mm nylon hose, but it needs to join into flexible pipe that's around 6mm (that feeds the boost gauge), and it looks like aquamist don't do an adaptor.

So I made one on the lathe!

Now I've had my mini-lathe & mill for a couple of years but rarely get the excuse to make stuff on them so this was a treat.

All pretty randomly made, and amateurish at best, but I've made a hose adaptor to join the 4mm hose to the 6mm hose quite firmly.

Here's a photo of this part next to a teaspoon for scale:

And the size of it against a steel rule (scuse the blurry phone pics!):

I was convinced it was/is going to snap, being made of ally and being so small (about 2.5mm wide at its narrow point) & hollow but we'll see in due course. I hope not!

All good practise and it's one little installation problem out of the way anyway, so I'm quite pleased with my handywork.

[Lara]

Your Avatar makes you look like a best friend I used to have, Can't think of his name now but his last name was Mann I think

We used to have great discussions

Lara

[Cret]

My friends tell me I look like him. I even have a scar in the same place.

Oddly enough, "Mann" is where I live (although some of my family are actually from Belgium).

Anyway - Last night I fabricated a bracket to bolt the water tank into the engine bay.

I already had a nice stainless water tank that my mate made for me ages ago so I welded the bracket on. It's going to fit just in front of the firewall on the left hand side as there is lots of space there and a bracket sticking out of the firewall with nothing on it. Perfect!

Now the pump is meant to be at the same level or below the tank to get a good feed. I've also welded a bracket for the pump to bolt onto as you can see in the pic. I think it should be ok as it's a tiny bit below the tank, but the good thing is that I have got a washer motor (attached in the pic) which can be used as a priming pump for the main ERL race pump.

Apparently, doing this really increases the flow through the pump as well, so I might need to get a smaller injector as the one I have got is 0.5mm and can flow 210cc/min before even having a primer.

I'm waiting for parts from Aquamist which they're still making at the moment so I can't use it yet, but I can get it all bolted in and wired up ready. I think I will set it so it only comes on at full boost, otherwise with the way a diesel uses its turbo the tank will be empty in no time!!!

[Troll Hunter]

Cret,

I've just come across this thread, and have read it with great interest. Have you actually got it up and running yet, and if so, what's your reaction? I'm intending to fit WI on my 300Tdi Defender. I've had the kit for some time, but have not had the opportunity to fit it yet. I got it from the guys at this link:

http://www.alcohol-injection.com/forum/install-pics/1997-mazda-mpv-2-5l-tdi-1382.html

This is about WI fitted to a Peugeot 2.5 diesel, and includes dynamometer results. Very impressive. I subsequently contacted the author, and he confirmed that he had not altered the injection pump in any way to get the "with WI" results. When I get around to fitting mine, I'll post the results.

Mike

[Cret]

Hi Mike

I've been meaning to update this thread for ages and have just been chocka busy and more than a bit sidetracked so apologies to anyone with an interest.

Yes I had the kit all up an running. The disco pulled slightly better with the water being injected but only under certain circumstances from the little testing I got to do. I think the jet was actually too large for my needs at 0.5mm and I didn't have smaller ones to try. Amusingly, there were a couple of occasions driving uphill under heavy load where there was visible mist in the rear view mirror!

Unfortunately having since that point been focusing on the subaru and getting the rebuilt engine back into that I've reluctantly had to sell the kit to raise a bit of cash. When funds permit I'd like to replace it with another kit and continue this as there was certainly some promise, but unfortunately I've been forced to spend the sale proceed on things like water pumps/fuel pumps/helicoil kits etc for the legacy. If anything I want a water kit for that now it's back on the road. I have a new ecu to begin mapping shortly and it comes with a charge temp sensor that can be logged so the water would be an excellent addition but maybe later when I'm not so skint.

Worth a try if you're considering it anyway. I definitely had some minor improvements, and that was without even any meddling with flow rates, jet size, activation pressure level etc so there are certainly improvements to be had. Just sorry I've not been able to expand on this further.

Jim

[Troll Hunter]

Jim,

Thanks for the update, and I'll keep checking back on this thread to monitor your progress.

I've just arrived in BC, Canada, from UK, and am still busily settling in to my new home, so all things Landy are taking a back seat for the time being. Added to which, I popped a front half shaft last Saturday, pulling some logs about on the property. New one on order, but will be about 10 days. Eventually, I'll get round to fitting my WI kit, and I'm hoping for improvements similar to those shown on the Peugeot MPV - see link.

Mike

[JimAttrill]

Sorry to join in so late, but I water injected my 300tdi last year. I used a boost switch and a mister from devils own in the States. I couldn't afford a fancy pump or tank so used a Disco headlight washer pump and the windscreen washer tank. Using 100% water (no methanol) I registered a 3% decrease in fuel consumption over quite a few kilometers. There was no apparent increase in power as the Tdi is not a good engine for increasing the fuelling when the WI switches on. Eventually the Disco pump gave up the ghost, I have another one which I can fit.

ps I worked on many piston and gas turbine engines which used WI. These were the RR Griffon in the Shackelton, the RR Dart in the AW Argosy and the Bristol engines (aka RR) Pegasus in the Harrier early mks 1 and 1A.

The WI was not used in the Harrier as 60 seconds of WI used the 55 imperial gallon of water/meth but also reduced the engine life as measured by a primitive computer to almost zero. As doing an engine change on a Harrier was not an easy task for various reasons (and the engines did not last their supposed 200hr life anyway) the use of WI was discontinued back in 1969. Maybe it was reintroduced later, but I somehow doubt it.

If you want to see the actual results I got with WI, I reckon you can search for them. I think it was all in the international forum.

[BogMonster]

200 hour engine life? Really?