TeamSwift

So, if we use a piggyback computer (like SMT6, fuel and ignition), we should get a good drivability down low. Thus at WOT at 1500rpm, we should get a nice increase in power and rpm without hessitations (can you confirm that please)? I ask you (and jardamuth) this explicity, because we have to choose a system (and spend some euro's ), and we don't want to regret our descision.

I have zero hesitations at any rpm, and Kevs car did not until it went cold again. The only hesitations I encountered were resolved on my car and Kevs by playing with the MAF signal (via piggy back fuel controller). Kev's hesitation in cold Calgary weather can be solved with some tuning. Both Kev and I use stock ECU with base timing set back to 2 degrees, and just a piggy-back fuel controller. I also don't run o2 sensor with stock ECU. I have tuned part-throttle on dyno and know enough to be able to change it if it is going rich/lean while driving. The corrections I need to make are very small in most cases.

Go to split second and check their prices on piggy-backs. You may find it cheaper than what you are looking at.

http://www.splitsec.com/

MAF placement may affect low rpm performance as well, but for the most part you should be able to tune it out with piggy backs (in my experience). Kev's car is nearly direct in line to turbo, mine mimicks the stock set-up by going around a bend before the intake of the turbo. The direct in line would likely be more problematic.

@ whattheeee:

You said it is probably a fail-safe issue (hessitations, because lot of flow/fuel at low rpms). This would still exists when we use a piggyback computer. When we use the piggyback computer to alter the MAF signal (increasing it), the ECU will go higher in its maps, and would inject more (at low rpm levels, the MAF could measure the extra flow when there is boost, so it would only be nessesary to get the fueling in the 12.5:1 range with the piggyback). But this way, it is highly likely that the ECU will go into failsafe mode (like you said). The only solution would be disconnecting the lambda (as you did), but this way, cruising will be very expensive , because the ECU doesn't alter the afr to 14.7:1. Because fuel economy is nessesary for me (80miles a day), I need the lambda and the closed-loop system .

So, would it still be possible to use a piggyback computer, run closed loop, and get the prober fueling all the way from idle till redline, without hessitations? If not, an standalone ECU would be the only solution, right?

I´m very interested in this topic...as i´m trying to turbocharge my mk1(http://www.teamswift.net/viewtopic.php?t=5324)...and i also live in europe and the smt6 and prs4 are praized here...

My doubt it´s exactly like ice said, would the smt6 be enough or do we need a prs4...translating "internationally", would a piggyback be enough or do we need a standalone?

I´ve been studying the prs4 and it looks very very good but it´s too expensive(900euros were i live)plus install and programing...and it doens´t work with our idle control...

What´s the piggyback to buy???? One that could put fuel and advance were it needed...

First off- most stand-alones are less problematic running fixed maps with no o2 sensor input. Some say they offer closed loop running for gas mileage, or wideband feedback and auto tuning but MOST are very troublesome and what is a good idea turns out to be a pain in the ass for the guy who can never quite seem to get his car tuned right. And trust me, with stand alones, if you, or the shop you go to doesn't know exactly what they are doing, you may end up stuck with some or many annoyances that the shop simply 'shrugs off' as part of running stand alones. There are some stand-alone systems out there that are capable of some very consistent, and rapid feedback but they usually cost many thousands more than the ones that offer the same feature that don't "really" work.

That said, as long as you know how it's influenced, and where you started, you should have no problems operating your Swift without o2 sensor and getting good gas mileage on more of a 'fixed' map. I think perhaps you guys are looking at this too 2 dimensionally. I do not get bad gas mileage, I get great gas mileage... 40mpg on the high-way. Why? Because at each rpm point it's not simply 'WOT' or closed throttle. There are many throttle positions at each rpm point. When I tuned my car, I verified load points on the highway, and simulated the load and speed with a seperate program on dyno to get an a/f of 14.7:1 while cruising around. This is why I get great gas mileage even though I don't have an o2 sensor running. Kevs car gets crappy gas mileage because we did not have the chance to tune these part throttle zones. I verified that his a/f was rich and safe and sent him on his way. If he hits even the tiniest amount of boost, he's rich. If we had more time that day, we could have taken the time to tune this problem out. So that even though he hits a couple psi even at light throttle, we could have made it safely leaner in that area.

So, what does this mean? Complete stand-alones are costly to set up, and expensive from software/parts. If you are going to run 7-8psi it is not a worthwhile expense. If you ask the right questions and do the right tuning, you can spend a lot less and still end up with a reliable and fun car. I learned the hard way and ended up with a very reliable set up on my car, and another.... with stock ECU, no 02, and no big chunk of change towards stand alone.

As for piggy-backs that I would reccomend... let me say that you need a piggy back that allows you to tune individual rpm/load points in true 3d. The SAFC is bad because it is either high or low throttle and it simulates the rest (simulated 3d... works great WOT, but doesn't allow multiple throttle setting changes). Systems like the Link or the Split second are true 3D which will allow fine tuning of various throttle and rpm positions.

I use 'mostly' the large zones (the big numbers) in the picture although you can see that there are over 144 or so smaller zones within those that you can manipulate if need be. This allows me to edit 1/4 throttle, 1/8 throttle, 1/16 throttle at one specific rpm point. This is good because it is my experience that running piggy-backs on our cars, you need fine tuning down low to get rid of the hesitation, and the rest you can generally make bulk changes.



There is a point where you will want stand alone, and you will start to lust after the extra tuneability, but for 7-8psi... it's just not worth it. I over doubled the horsepower of hte stock motor with just a piggy back, managed better than can be expected gas mileage, in a reliable daily driver without too many quirks.

Do expect some quirks to come up regardless, just don't think that stand-alones solve them... they generally create more 'quirks', especially the wrong system in the wrong hands. If you can keep the factory ECU, keep it, it simplifies your life a whole lot.

@whattheeee: Thank you VERY much for your answer. It helped us a lot. Normally I do think in 3D, but this time.... We could set the mixture at low throttle to 14.7:1 open loop. This way, gas mileage will be ok. Stupid I didn't see that before . We have to pass an emission test in our country at idle and at 3000rpm. CO has to be below 0.5. I think this would be 'easy' when we tune it ok. If not, we always can connect the lambda, so the ECU can use closed-loop to get 14.7:1.

So probably we will use a piggyback computer (as you said, it not worth the hassle and money to use an aftermarket with 7-8 psi). But you recommended the Link or Split Second. The link will cost about 450euro's in the netherlands without the communication cable (which is also unreasonable expensive). With the Link and/or SS, ignition cannot be retarded (only fueling can be altered). The SMT6 is both ignition and injection.

What do you think, based on your experience, logic, etc, is this a good device? I've heard that the SMT6 doesn't interpolate between map entries and I know i doesn't have many map entries. I would prefer this device because its cheap (300euro) and also does ignition (when there is no boost, we want to use 7 to 8 degrees BTDC (RON98) for ignition). Maybe you know someone with a turbosetup and a SMT6 computer? I would like to ask for his experience with it.

Again, thank you very much for your help!

PS: If I understand correctly, the standard ECU gets confused because of its lambda sensor when there is boost at low RPMs. Disconnecting it (just let the lambda input to the ECU hang in the air?) would solve this problem.

I think this thread is a perfect example of the true value of this comunity. There is a lot of money worth in informations, hours of testing and costly mistakes avoided by reading this post alone.
I hope everyone here take a moment to realize and aknoweledge that fact.

On a side note, we're still looking for supporters for the teamswift funds so we can make the site even better and last longer.

Thanks for you all.

Jess

_________________

I completly agree with Jess, this post is becoming very interesting and Teamswift it´s a HUGE resource site...

Regarding the smt6 it seems very promissing and Whateee completly convinced me...piggy-back is the one to go(in my case)...

But there is a little problem in my case, my project is a mk1 and it has a vaccum actuated advance in the distributor...i think that it´s not controlled by the ecu but the doubt still remains because there 2 little wires coming from the distributor to the ecu cabling...a friend mechanic told me that this wires are an input sensor because this car doesn´t have a crankshaft sensor and uses the distributor to signal the ecu for the engine position...

Can anyone confirm that?????

you should pm OZmidnight if he doesn't respond to this on his own.

_________________
'92 Swift GT, '94 Swift GT, '88 Turbo Sprint, '98 Swift ... all sold years ago

The Mk1 has a standalone distributor...ECU controls fuel only. My distributor was modified in the old-school way with a very different advance curve to suit the cams, etc.

As for ECU's, there are a host of very good ones on the market....Link, Haltech, Motec, Autronic, Wolf, Hawk, EMS etc etc etc. They have all proven their worth on both street and competition cars here. Some of the afm type adjusters (not sure about all) are limited by the map limits in the original ECU.

As for piggybacks, the Link and the Unichip (not really a "chip") would be the two I'd recommend from listening to people that have used them. A Unichip, fitted and tuned on a dyno here, is in the vicinity of $1200Aus.

Hope this informative?!

_________________
Don't ask me, I don't know anything. And just because I'm a senior member, it does not mean I have as many wrinkles as ellpee. 1500+ posts and still I haven't got a clue.

Split second and LINK both have timing/fuel combo controllers. The Split second unit should be cheaper.





Don't know of anyone whose used the Perfect power off the top of my head. I like the idea for datalogging and things like that, although I can't say how well it would work or what the actual problem would be with the 'interpolation between cells'. Sometimes tuning is a problem more than people realize, and sometimes you should in reality be able to feel a 4-5 degree change in timing if there shouldn't be that big of a change in that spot. With a piggy-back, I'd think it difficult to screw it up too badly, but the only way to really find out how good it is would be to try one. It has some good ideas, and it's cheap.

We have used almost every type of engine management at the shop, and there are quirks with all of the ones Ozmidnight mentioned.... Some of the Motecs are very quick and nice, but probably at a much higher cost than the others on the list.



Well the way the ECU functions is that it no matter what throttle position, after 4000rpms it goes into open loop, fixed map. At the same time it should go to open loop straight fixed map under 4000rpms and full throttle. It seems the latter is not neccessarrily the case though as you heard Jardamuth comment that he hasn't seen any stock ECU turbo Swift rich before 4000rpms (until now. ). I'm merely suggesting that the o2 sensor confuses the ECU before 4000rpms as it seems to fit within the logic of econo-box meant for fuel economy factory tuning. All I know is that on Kevs car and mine, we were run without o2 sensor and you can see the wide-band on the dyno shows rich. Draw conclusions from there...

Thank you all for replying! Now we know a lot more

Currently I'm looking at the Unichip. It has fuel and ignition, and can be installed and tuned on a dyno for 710euro. If the unichip is the solution for us, we go with that.

The reason we want to tune it, is because yesterday we removed the head to check the headgasket. Well..... we cried ... A hole in the piston and a lot of knock damage in the third cylinder (in the block and in the head). We need a new head and bore out the cylinders to fit 1.6 pistons. We maybe can get hold of a GTi engine (complete with ECU) for 200euro. Obviously we hope it is still for sale.

The engine was running way to lean under boost (our fault). Luckely my engine does run rich enough under boost, so I don't have any damage (we checked the AFR while driving in my car).

We will post some pictures We hope to get some answers at the end of this week, so we can fix his engine, and dyno tune both cars (hopefully with a unichip).

BTW: I've been looking for some info about the unichip, but can't find enough. Things like:
- Is it 3D (ignition and fuel)?
- Can it use MAF for load sensing?
- Can it use a seperate MAP for boost detection, so ignition can be retarded when there is boost (x degrees/PSI)?
- Does the unichip comes with a serial cable and software so we can look at (and change) the parameters/cells of the unichip?
- Can it control boost (using a bleeding valve)?
- Does it interpolate between cells?
- etc...

If someone knows the answers, please post them (or a link to all the details of the unichip).

Biggest problem with unichip is that it doesn't come with cable and software:( Only workshops approved by Dastek can tune unichip (got software). SMT6 has free software, so You can tune it by Yourself. Also smt6 has 5th injector control, also it can be used to control water injection or other devices.

_________________
Mk3 GTi Saudi Arabia edition.

You're right. I just phoned 'my tuner' and he cannot provide the software and cable So, when set up, we can't make any changes anymore ourselfs.

Do you think the SMT6 is a good replacement for the unichip? And my biggest concern is; can the SMT6 use the MAF signal as load sensing (MAF vs RPM)? I don't want to use the TPS for load sensing (obvious reason). Because almost nobody has experience with the SMT6, so I'm a bit scared to buy the wrong device... Maybe someone who has (extensive) experience can post some results/experiences in this topic?

We have 2 smt6 in workshop, so soon those will be tested (altought boss says that those are shit and he has built better:P) in eclipse 2g gs-t and peugeot 405 mi16. I'll ask about it tomorrow if it will be possible.

_________________
Mk3 GTi Saudi Arabia edition.

THis thread is definitaly getting more and more interesting.



I must add that I don't run them with disconnected O2 sensor as they're set for daily driving. I myself use the pocket rocket for 400kms+ trips and having the lambda disconected would give overkill gas milleage.

_________________

Ok, we've made the descision to go with a piggy-back computer. We don't know which one yet.

But before we are going to take the cars to the dyno-tuner, we want some confirmations from you guys so we know for sure we are doing te right thing

So would you be so kind the confirm the following things:
1) We are going to run 8.5 psi at standard compression with RON98 for fuel. Base ignition will be the same, but for every psi of boost, ignition will be retarded with 2 degrees. Is this ok/safe?

2) Can we use the standard MAF for 8.5psi, or is it wise to get a bigger one (nissan SR20DE(T))? If standard will measure at max. 7psi, we would get a rich spot because fuelpressure must be raised to get 12:1 at max boost. Right?

3) We will disconnect the lambda, so ECU will run open-loop. The piggy-back will be setup to run lambda 1 (14.7:1) at part throttle and below 4500rpm. Ok?

4) At 8.5psi we would get around 170 a 180hp and a torque of 160lbs/ft or 215Nm. Right?

5) We will buy a clutch from ACT which has 53% more clamping force and a max. torque of 177lbs/ft. Sounds ok?

6) The engine will run reliable with 8.5psi and 2 degrees retard/psi of boost for a daily driven car? Right?

7) At how much pressure must we set the fuel so we can compensate for the extra air? Standard is about 2.5 bar (37psi). What do we need for 8.5psi of boost? 3 or 4 bar (44 or 60psi)? Just an indication is enough

8 ) Any remarks/tips/hints for us before we tune the car?

I hope we can get some confirmations/answers so we know for sure we are on te safe side. Thanx!

So would you be so kind the confirm the following things:


At 2 degrees base timing, the top end timing maps really are not that bad. Real-world... I've measured like 22-23 degrees at the crank above 4000rpms under full load on the dyno. I would consider a timing controller to change lower rpm-mid-range timing but I don't think there is much power being lost in the top end with the base timing set back. For 7-8psi, it would be nice for lower stuff, but neccessary by any means.



You can use standard MAF... the voltage will max out early on, but it still moves up in the rpm table, and provides enough fuel. It's somehow almost magical, but the only problem you'll run into is if you manage to max out the injectors... which you shouldn't be able to do with 7-8 psi. After a point you may need to start tuning with fuel pressure or doing it smart with larger injectors).





part throttle below 4500 rpm sounds good. If you have boost in an area, make it a little richer. Not quite as important to be pig rich with a few pounds of boost in lower rpms.



At the crank perhaps. You should be able to get 150+whp around there if all your duckies are in a row. If you are porting the exhaust side of the head (something I reallly reccomend).




ACT clutch should be good. A good stock replacement should hold the power, but not neccessarilly between really hard shifts. I use an overkill pressure plate I picked up from Adam (runs the webstore) and a replacement disc. It's a weee bit heavy and if you go with it, there will be certain neccessary steps to take to prevent hassles associated with that heavy of a clutch plate




Just retard base by 4 degrees and call it good. If you retard 2 degrees/psi you'll end up running like 10 degrees of timing under boost after 3500rpms in some areas. Not super fun.




Use the air/fuel calibrator to change how much fuel you need at given point. Raise the fuel pressure by 4-6psi to start. If you need more because of AFM being maxed out, and your injectors maxed out, turn up the pressure a bit. It's not really a safe fix for a daily driver, but you 'shouldn't' need different injectors at 8.5psi of boost.



see above replace the stock fuel rail at minimum (if you haven't), it's restrictive and otherwise better in the garbage. You may want to consider replacing the fuel pump, although that will depend on what set up you end up running.

Split second sells their timing and fuel calibrator for 420 Euros. Just the fuel calibrator is more like 170 Euros unless you go with another product they have which is a fuel calibrator but you can use a larger GM maf with it and it will send the signal properly (how well this one works... I dunno... never tried). Downside is price will go up to neighbourhood of the timing/fuel calibrator. It's that or try the SMT6, or something like Greddy EManage, or.... whatever... all you really need is fuel control IMHO.



unless it was tuned for 14.7:1 on say a dyno, thus resulting in 40mpg right?

oh, one other thing... seeing as you are going to run it as a daily driver. If you drive like a maniac or if you are planning on doing any auto-xing, I also suggest tuning part throttle maps in the high rpms. There will be some lean spots higher up in the power band when you are trying to modulate the throttle around a corner. Producing big boost, and being in the wrong part of the fuel/timing map in the ECU with half throttle isn't where you want to be for reliability.

For WOT/closed throttle, it's pretty easy to set up.

Ummm, it's a lot more reliable to have a unit capable of building a load map with inputs from both the throttle position sensor and the air flow meter sensor. Generally they are there to modify to the air flow meter signal to adjust your fuel and they just simply read the TPS to determine a position on a map for you.

Thank you very much for your extensive answers! We will tune part throttle until high rpm, so that must me ok.

Btw when I asked for 170 a 180hp, I meant crank hp, not wheel

Would this work? Leave base ignition at 6 degrees and when boost is detected (>1psi), switch to base ignition - 4 degrees. This way, we would have nice offboost drivability and safe boost ignition numbers. Right?

Thanks again for your answers!

tune full throttle first so that is safe, then work from there.



Well, it is still possible depending how many changes you make. Some exhuast side port work, different intake manifold and you'd be surprised what you can accomplish.




Exactly what I would use it for... play around advancing timing for better off-boost performance. I run 2 degrees base, but like I say, it'll depend on your car and what you put hte motor back together with. You may find in that area more or less. There is one spot in the mid-range 3500-4500rpms somwhere that I would have loved a timing controller for. Something in that range needs adjustment.

Uhhm the smt unit comes also without a cable or you'll have to get your hands on a develpment kit. Cause if not your tuner will tune it and lock it.

I had the smt5 unit. I made my own cable and had the guys where I bought it from here in holland to give it to me unlocked. I had it hooked up with a map sensor to tune my car while on boost.

P.s. The SMT5 unit is dead now. Don't know what happend it just died one of a sudden:?

We've completly damaged the engine. So I'm gonna buy another original one tomorrow.

Leave everything so if it's in good shape.

Here some horror pics for you all... There is even a hole in the 3th piston

for the SMT-6 that Adam & Phillip were posting about ..

_________________
'92 Swift GT, '94 Swift GT, '88 Turbo Sprint, '98 Swift ... all sold years ago

uh. nice space green coolant, what the hell kind of antifreeze is that?