Hi guys

I'm building a high boost SPI turbo with a second piggy back controlled injector and have a question about fuel regulation. Most of the set ups I've read about seem to use a aftermarket regulator for the second injector and retain the original regulator for the primary injector in the throttle body. Surely if you use the standard regulator on the primary injector as the boost increases the fueling will be diminished by the pressure in the plenum and vacum feed for the regulator. This meaning you will have to overcompensate with the second injector? The only way I can see around this issue is by using 1:1 boost increasing regulators independently for both the injectors and disabling the original unit by removing its guts. Apart from the cost of using 2 regulators is there any reason why I shouldn't be considering this? Can the original regulator cope with the added boost with out a boost increasing regulator? I was under the impression N/A regulators rely on manifold vacuum where as turbo regulators rely on boost pressure.

Cheers

Ben

Have a squatch at the turboing burple thread for info.

Cheers

D

I'm reading through it as we speak but I don't see how if using the standard regulator on the primary injector it will be able to fuel correctly with positive pressure. For example I'm under the understanding that if you had 1psi in the throttle body you would need to pump 2psi though the injector to get it to squirt 1psi of fuel. There is 1psi back pressue against the injector in this situation no? This is why they use boost refferenced/increasing regulators (generally 1:1) on lots of turbo charged engines. The standard regulator in the throttle body is for a N/A engine and will not increase with boost. I guess you can adjust the timing on the second injector to to compensate the fueling as the pressure increases but I would of thought having a boost increasing system on both injectors would give better results and mean a more consistant fuel map?

Cheers

Ben

I don't see where you are getting the bit about the boost reducing fuel pressure.

The plenum is positive pressure and the standard injector is releasing its fluid into that as it passes through the throttle body.

The second injector is mapped so that there is additional fuel when the amount of air exceeds the amount of fuel from the first injector.

Cheers

D

My brain has been a bit squidged by the amount of contradicting articles I've regarding turbo charging a N/A over the last few weeks. I think using a rising rate fuel regulator is only necessary if you are try to squeeze more out of an existing injector than it is actually supposed to provide. Since I will be using a second injector with a boost/MAP sensor input and programmable ignition retarding I can't see why I would need it. However it seems most regulators are using a vacum or boost supply to adjust their regulation. Regulators for turbo charging use boost to increase fuel pressure and N/A regulators seem to use vacuum. Some seem to be capable of both most at a 1:1 rate. If the original regulator in the throttle body was designed to use a N/A vacuum signal to increase its pressure I am wondering how this will respond when the the throttle body contains positive pressure. I guess this will no longer be an issue since the second injector will be supplying the majority of the fuel which can be adjusted by the map. I believed injector fueling was just controlled by pulse width modulation and if that was the case why would regulators have a boost or vacuum input? I can only think to overcome pressure in the the inlet? I also see that Smokeme always recommends the use of a 1:1 boost increasing regulator? I don't know where I got the analogy for needing higher PSI than the boost pressure being required but I certainly read it somewhere. Of course that does not mean it is correct.

Cheers

Ben

the spi unit is a steady rate regulator... it does not change at all.
so... if the regulator is set to 1bar and you are boosting at one bar, nothing will come out the injector (once over one bar air will be forced thru the injector)

what ever way you chose to do it wont matter as once tuned it will all work as long as the engine gets fuel. in my opinion it would be best for the engine to get most of its fuel from the spi unit injector and the extra required fuel from the external injector, but it seems most work the other way around... but it is the cheaper way.

mpi is the best way to fuel

Thanks Craig I was starting to think I was going mad. I would certainly consider going MPI with the punto 75 motor but getting one where I live won't be the easiest thing in the world since we have no proper scrap yards or breakers. I'll certainly keep a look out for a 75 though. Getting a MPI head shipped would be expensive and not worth the effort unless I go the whole hog. Since ultimately you end up with a car worth 2-2.5k I'm not prepared to put in that type of investement at the moment. Also the local authorities are pretty strict on modifications so I'm hoping to put it on the log book as a 1.1T. (Worked with my old turbo technics 205 1.9T)

I have 3 SPI 1.1 engines available to me also all with around 35,000 miles which need to be used.

If I use a 1:1 boost increasing regulator on the primary injector untill it's maxed out and then supply the remaining fuel with the second injector this would make sense then.

Cheers

Ben

My 5 cents:

I was also wondering about the exact same things as you do, and no, you are not going mad, but Craig already explained. What I want to add is the other reason for fuel regulators (apart from repeatable performance at a given air pressure and naturally varying fuel pump pressure) is (my guess) to make the injector behaviour linear, so that doubling the pulse width means doubling the fuel. It makes programming ECUs much easier (for example temperature sensors are usually not linear and there is a fair bit of code in the ECU to deal with that).

BTW, where are you from then?

Hi Woj

It's nice to know I'm not the only one thinking this. Letting the original injector get squashed to nothing more than a dribble during high boost seems like a less than ideal solution to me. Like you say I think it would probably make programming/tuning more linear and in that sense easier. I was originally just going to set the car up with a wideband from the passenger seat but unless I'm 100% happy with the fueling it will be going to a rolling road. Also I may consider putting a lambda bung on cylinder 4 just for set up as it has a reputation for going lean due to feeding the brake servo. Perhaps the servo feed can be relocated in a way that it gets an equal feed from all cylinders. Has anyone done this?
I live in Jersey Channel Islands by the way.

Cheers

Ben

Honeymonster said:

I was originally just going to set the car up with a wideband from the passenger seat but unless I'm 100% happy with the fueling it will be going to a rolling road.

Click to expand...

Getting the fuel right with WB is relatively easy and you will be happy with the result, at least that's what I did and I am. It only takes a while longer than you think. Getting the ignition right on the road is a whole another story.

Also I may consider putting a lambda bung on cylinder 4 just for set up as it has a reputation for going lean due to feeding the brake servo. Perhaps the servo feed can be relocated in a way that it gets an equal feed from all cylinders. Has anyone done this?

Click to expand...

It shows that you did a fair bit of research, but then:

1. I understand why it would go lean in principle, although it sounds dodgy to me. You suggest that it gets extra air from the servo? The map sensor input is just next to it, so it would show on the map anyhow, so ECU would accomodate this.

But anyhow, you primarly want to make sure you stay rich on boost, and there is a oneway valve on the servo (has to be, otherwise you would be boosting the air into it), so no problem.

2. Where exactly do want this lambda bung, on the exhaust manifold on cyl. 4? It will kill a WB sensor very quickly, way too hot there for it. On mine it says it should be at least 12 inches after the turbo.

Hi Woj

Sounds like you've done a lot more research than me.
I know SPI is never going to distribute the fuel as evenly as MPI and had read that cylinder 4 is the favourite. This may be completely unrelated to the brake servo as it was just something I've picked up while reading various threads over the last few weeks. If there are any other turbo owners reading who have melted pistons can they confirm that cylinder 4 is the most common? You're probably right about positioning the sensor so close to the turbo. It was just an idea that popped into my head which I haven't really looked into in great detail yet. Of course it would only be a temporary thing. Maybe even a cheap 2 wire lambda (that cound be sacrificed) would give a good indication of whever or not it was running lean on full boost. If there is a common cylinder with a problem it would sure be an advantage to get some readings from it.

Cheers

Ben

OK, I missed the fact it is an SPI, it all makes sense now. Yes, it is a (well?) known fact that it runs leaner on the outside cylinders actually, both 1 and 4, and this is natural for the SPI engine if you think of it. But I am not sure how leaner does it get on boost, and I am not sure too if any SPI turbo owner ever managed to measure this. I wouldn't know at all as I run MPI.

Anyhow, only WB lambda is useful here, but it will burn when mounted in the exhaust manifold. Narrow lambda may survive this (don't know much about temp resistance of narrow lambdas), but is useless, it will show you above / below 14.7 AFR, and this is not enough on boost to know, 14.5 is too lean for boost and narrow will show rich...

Cheers Woj that's useful information. I would be keen to go MPI but like I said before I will struggle to source a head. Secondly If I did no doubt I'd be going for a full stand alone ecu/management system. Of course my piggy back ecu would support 4 injectors meaning I could use SPI for off boost fuelling but this would seem less practical to me. Going stand alone MPI would mean I'd only need to use a single 1:1 boost increasing fuel regulator which was my original issue at the start of the thread.
Going lean on any particular cylinder in this case would also be very unlikely as the overall mixture would no doubt reflect the combustion of all cylinders when taken just before the cat. The more I talk about MPI the more it makes sense but I'm trying to do this without spending an extortionate sum of money. I know ultimately a good cinq turbo will not be worth more than 2-2.5k. This is just a little project for me to tinker with on my days off. I've always liked the character and size of the cinqs. Don't get me wrong I'm not planning on selling it but I've said that about the last three projects I've completed and sold on at a loss.

Cheers

Ben

if i remember right the spi setup is as so...
fuel feed from pump goes to the extra injector then to the rising rate fpr (which keeps the pressure in the extra injector) the fuel fuel out of the fpr then goes to the spi unit... once out of this it goes back to the tank.

if you simply move the fpr to the end (so on the exit back to the tank) both injectors will be rising rate. the spi unit will never drop bellow its set pressure even with full vacume unlike the rising rate regulator.
surely that will give you the best of both worlds (spi unit wont turn into a dribble while on boost)

Honeymonster said:

Cheers Woj that's useful information. I would be keen to go MPI but like I said before I will struggle to source a head. Secondly If I did no doubt I'd be going for a full stand alone ecu/management system. Of course my piggy back ecu would support 4 injectors meaning I could use SPI for off boost fuelling but this would seem less practical to me. Going stand alone MPI would mean I'd only need to use a single 1:1 boost increasing fuel regulator which was my original issue at the start of the thread.

Click to expand...

Actually, Tricker did this in his turbo in DK, look for his threads: SPI for off boost, piggy back on MPI for on-boost. Waste of good running MPI power off boost if you ask me, but he is a very experienced tuner and he was very happy with the setup and results.

... without spending an extortionate sum of money.

Click to expand...

Serious and honest advice from the heart: abandon now! It gives a great pleasure and satisfaction to build a turbo setup, I am constantly on high since I made mine run in June, but budget wise it was a disaster, worth a decent second hand card. I basically overshot the assumed budget twice, and now that I looked at it, it was unavoidable. And I am not yet fully finished: proper dyno session still ahead of me, as well as some body shell work.

craigb244 said:

if i remember right the spi setup is as so...
fuel feed from pump goes to the extra injector then to the rising rate fpr (which keeps the pressure in the extra injector) the fuel fuel out of the fpr then goes to the spi unit... once out of this it goes back to the tank.

if you simply move the fpr to the end (so on the exit back to the tank) both injectors will be rising rate. the spi unit will never drop bellow its set pressure even with full vacume unlike the rising rate regulator.
surely that will give you the best of both worlds (spi unit wont turn into a dribble while on boost)

Click to expand...

I can see what you getting at Craig but how will you set the initial pressure on the extra injector at the start of the loop. As soon as you start restricting the main fuel return I expect pressure will increase in a linear fashion on both injectors. This is great if you want to supply both injectors with same amount of fuel but I was under the impression the second injector needs to be significantly larger requiring more pressure? Of course you can't wind both pressures up as this will over fuel off boost. Maybe you can overcome this by keeping the first injector running which previously it wasn't. I'm not sure but most people seem to be using 1-1.4 bar on the primary and 3 bar on the secondary but this is usually without rising rate. I don't know what the size of the standard injector is but I'm guessing 2 of these maintaining 1-1.4 bar above inlet boost would be insufficient so you would need to find the perfect second injector happy at 1-1.4 bar.

Woj I know that everyone on here who has built a turbo seems to have gone wildly over budget. I may do too but I think I come up with a spec I'm happy with for a decent amount. I'm lucky enough to have good friends I share a workshop with. This includes welder fabricators, machinists, motorcyle mechanics and a panel beater/painter. I do electronics myself so except he rolling road we have pretty much got it all covered. We are also lucky enough to have a car lift.
The biggest expense will be suspension and brakes since most of the other bits have been purchased cheaply as a second hand kit including a good quality pggy back ecu. The engines were almost free since all the cars were purchased for less than £50 each. I just want to make sure I understand the the project before I make a start. The first thing that seemed wrong was the fueling which is why I started this post. I'm looking for 7 psi on standard pistons using an oil cooler, intercooler and a 1.5mm decomp plate. If this melts I will simply fit everything to another engine and try it slightly lower.

Cheers

Ben

Honeymonster said:

Woj I know that everyone on here who has built a turbo seems to have gone wildly over budget. I may do too but I think I come up with a spec I'm happy with for a decent amount. I'm lucky enough to have good friends I share a workshop with. This includes welder fabricators, machinists, motorcyle mechanics and a panel beater/painter. I do electronics myself so except he rolling road we have pretty much got it all covered. We are also lucky enough to have a car lift.

Click to expand...

Well, you've been warned Indeed having for example welding for free is a huge save, I have spent on mine a fortune, but at least it is a damn good welding job.

The first thing that seemed wrong was the fueling which is why I started this post. I'm looking for 7 psi on standard pistons using an oil cooler, intercooler and a 1.5mm decomp plate. If this melts I will simply fit everything to another engine and try it slightly lower.

Click to expand...

That won't melt, should be no problem. It is actually close to my spec, I blow 0.8 bar max (this is more than 7 psi), oil cooler and IC, and the decompression is similar, also with a plate (different size, but I also used different approach with HGs than most folks here). Having proper ignition settings (which I just found out I do, cheers Craig ) this keeps me very cool, my EGT hardly ever reaches 700 degress on 98 fuel. Truth is though I haven't pushed it all the way up the revs yet, but I consider these results so far far from melting point.

Fueling / pressure wise I fully understand your concern (guru Aaron here says fueling is of "outmost" importance and he is right), and this is why after being puzzeled about this myself for a while, I decided to go MPI and single management system (regardless of what it is in practice) that lets you set fueling and ignition through the whole load range. This is something for you to keep in mind: to get it running properly and safe you need to work out the ignition in the whole load range: off and on boost. Not sure how good of a job those piggy back systems do here, looks a little bit hellish job for me to set all this up.

woj said:

This is something for you to keep in mind: to get it running properly and safe you need to work out the ignition in the whole load range: off and on boost. Not sure how good of a job those piggy back systems do here, looks a little bit hellish job for me to set all this up.

Click to expand...

My piggy back system can retard or advance the ignition in 17 positions through out the rev range IIRC but setting it up is another matter. The Let Rip software seems pretty intuitive but it's probably best left for the chaps at the rolling road to tweek. So whats this secret head gasket system them?

Cheers

Ben

Honeymonster said:

So whats this secret head gasket system them?

Click to expand...

It is no secret, I described it my turbo thread, two metal (two-layers each) head gaskets, one below, and one above the decompression plate. That's it. The decomp plate needs to be thicker in this case because two metal HGs still do not add up to as thick as a single regular HG is. Plus I needed even thicker decomp plate, because my head was skimmed more than I anticipated. This setup gives you a bit more robustness with probably a little bit less safety margin, the downsides are a) risky tightening procedure (to my understanding on the endurance edge of the head bolts), b) both block and head need to be skimmed for this for everything to mate properly.

Hi Woj

Sounds like a good system you have there. No doubt it should be reliable but surely to get the block skimmed meant you had to totally strip the bottom end of the motor and get it machined? Not a massive task since no doubt you changed the shells anyway but the engines I have are low mileage so I wasn't planning on pulling the pistons. Then I take it you had to add the amount you got skimmed off the block and the head to the decomp spacer you had made up. Secondly like you say you'll nail a piston long before you pop the gasket but with MPI, wide band and the knowledge you clearly have I doubt this is a major worry for you. Certainly if you don't melt a piston you'll be spending a lot less time changing head gaskets than most turbo owners. You'll probably be able to run a bit more boost too although best to try not to.

Cheers

Ben