son-of-frimbus

I have been reading this forum since I purchased my Xb in December '06 and I too would like to turbocharge my car. I have debated purchasing a kit to do this task, but I have found that kits are not always "drop in" and thus I have decided to design my own. I have done this before with 2 Fieros that I own. I have learned quite a bit by reading this forum, but I have noticed that quite a few people have had to reset thier ecu's from time to time due to the learning characterics of the ecu and I would like to ask a couple of questions and see if anyone has tried these possible solutions.

Has anyone ever tried to use a calibrated MAF pipe? I see that it looks like ZPI had some success with this. I have 365 cc/min. Denso injectors and by my calculations, assuming a stock MAF is 55 mm and using 8.5 PSI boost the re-calibrated MAF pipe should have a diameter of just over 71mm.

Has anyone ever tried to use a voltage dividing circuit on the oxygen sensor to "scale" down the voltage. This would report a "leaner" signal to the ecu and the ecu in return would would give a "richer" signal to the injectors. It seems to me the ecu is using (in General Motors terms) the BLM, or block learning multiplier to do long term adjustments to the fuel map and that might be possibly why guys are having to "reset" the ecu every couple of days. The voltage divider circuit could be a variable resistor (a potentiometer) that is connected across (in parallel) the supply voltage (+5 volts, I assume) and the output to the ecu on the oxygen sensor.

I have a Garrett T-2 (I have had very good luck with this type of turbo over the years) ready for the Xb, but I have 29 months or 30,652 miles to go before the warranty is up, so it will be a long time before I drop anything into my engine compartment. I look forward to your opinions.

arizonaturbo

iTrader: (1)

I understand what you are saying about the MAF. We did this quite often in teh SR20DET world by using the Ford Cobra MAF on the FWD SR20's and using the larger 300ZX TT MAF on the RWD SR20DET's. Most of the Nissan MAF's top about ~5.12v so rewiring a new MAF harness and running the larger MAF allowed us to utilize the large amounts of comressed air.

The o2 sensor was usually not a problem on the OBDI cars, but this did create problem on the OBDII wehicles. Most of the time we had to deal with the check engine light.

I own a 2006 xB also and will attempt to install the Greddy Kit on this vehicle. I know Rollhard installed 310cc? injectors from a Celica GTS only to find out that the e-Manage was unable to cut the duty cyle low enough for the car to idle. I know this is a common problem with Hondas and Nissans from past experience. Honda's usually don't like 310 - 370cc injectors at idle using the stock ecu. The most common cure for this is to get a remapped ECU. Nissan has JWT where as Honda has Hondata or various other stand alone applications which will sole this problem.

The Garrett T series turbos would be a perfect match for our 1.5L motor. The Nissan T25 and T28's usually can generate 225-250 whp safely with 8-12 psi on an SR20. In extreme cases the Pulsar GTiR T28 can produce close to 300 hp.

son-of-frimbus

I'm glad to see that someone has tried the MAF experiment, but have you thought about trying on your Xb when you put in the Greddy kit?

Looking at the part number for the Xb MAF it looks like it is used by even the 4.7 V-8 as well as several other engine/platforms and looking at the airbox (I purchased an extra one off Ebay for research purposes) it looks like the MAF plugs into a tublular section of the airbox that has the laminar flow grid as part of the tube section. I was figuring to modify the stock airbox by changing the diameter of the tube and flipping it 180 degrees to face the firewall and turbo. The TRD turbo Xb has an airbox top that does exactly that, but Ican't see what they have for a MAF pipe muchless what the part number is for the airbox top. Wish I had that number for X-mas.

OBDII cars use a 4 wire O2 and update the BLM's much faster, I heard that the Honda's use them for closed loop idle in real time mode, but I think that if one were to use a voltage dividing circuit, one could "fool" the computer into thinking that the exhaust stream is excessively "lean" and the computer would "enrichen" the mixture accordingly. I think that it is possible to eliminate the "check engine" light with this fix.

I know I might be pushing the lower limit on the duty cycle for idling with the 365cc's Denso injectors, but I got them for $10 a piece on Ebay, so if they don't work I won't feel so bad and get some 280cc. What are the factory injectors rated at anyways some people here have said 210 others said 280 ... I have a hard time trying to get this info. I calculated 215 based on .6 BSFC and 108hp with 80% max duty, but it's only a guess.

I am a little scared about using a piggyback, but only because I have no knowledge in this area. On my Fiero, I can hack the ECU using Moate's AutoProm and Mansur's TunerPro software. I wish I could hack the Xb's ecu, but I haven't found anyone who has broken the factory code - yet.

I was going to go with a GT2052 for journal bearing and for ball bearing the GT2554R, but I saw a brand new Sunbird GT turbo going for $250 and it's turbo map very similar to the GT2052, so being cheap - I bought it. I was going to fab the manifold out of stainless buttweld pipe and tube fittings and machine up some flanges.

Is the Greddy a top mount manifold? From laying out the turbo in the engine bay it looks like a side mount (like the TSI kit) would put the turbine housing really close to some of the firewall stuff, like an inch or less on the T2 unit I have. It looks like the TRD version is a top mount, but it's hard to tell from the picture on the TRD website. Maybe it's better for top mount on our cars - I don't know, but I still have some time to decide.

Thanks for the reply arizonaturbo!

arizonaturbo

iTrader: (1)

I am sure your voltage dividing circuit will work with the bored out MAF. I am not that technical when it comes to electronics, but the theory seems correct.

THe stock injectors are somewhere in the 210-230cc range I believe. The e-manage can olny control injectors up to 50% larger than stoock with the optional injector harness. I would say 370cc or 36lb injectors would be the largest I could run with the e-Manage.

I do know that Toyota ECU's have always been some of the hardest ECU's to crack the code. I do know that the Supra guys widely used the HKS GCC and VPC units prior to the AEM EMS with good results. I am interested to see how well the e-manage will work on the 2006 xB ECU.

Have you considered the S13/S14 SR20DET T25 as an inexpensive option? You can easily pick one of these up for $100.00 - $150.00. You can easily fabricate a log style manifold using 8ga steel or stainless steel elbows. No need for an external wastegate since these turbos both have actuators set at ~ 7-8psi.

I haven't really seen how the Greddy turbo and manifold mount up. I opened the package, took the turbo and manifold out, took the turbo apart and sent the turbine housing, heatsheilds & manifold to get HPC coated and I had the compressor housing polished also. I am waiting to get another o2 bung welded on the downpipe for a wide band o2. Once this is done I will get it HPC coated as well. I pondered sending the turbine housing to Extrude Hone, but figured it wouldn't help much on such a small turbo.

lastlookcustoms

The next one that I do, I am going to be running the MAP-ECU which gets rid of the MAF totally, converting to a speed density setup. The ECU offers full fuel mapping as well as MAP enrichment tables all controlled by a laptop. We have used this setup in another car with great results. I think this simple piggyback will solve all the scion computer issues

arizonaturbo

iTrader: (1)

This just came in from Greddy about the 2006 xB:

We are currently working on a newer kit for the 06 model there are a few other things the 06 needs besides just rewireing the tune is different and there are a few adapters for the Emanage .

rollhard

I have the 310c injectors off a celica and what you will have to do is trim the airflow correction down and 0 out the low throttle/air flow voltage from the base Greddy map. If you are doing your own kit, good luck. In most cases, it might cost more than the $2500 for a Greddy kit. I wouldnt worry about reclibrating the AFM though. Its not going to matter because the ECU wont be able to recognize the amount of boost going into the motor anyways. Your intake draw wont necessarily by the boost amount. I have tried different things, in an attempt to get the right combination for you guys. I will be installing a Walbro 255 this week and doing no additional fuel correction other than negative on the airflow to trim fuel down. Ill see how that goes and let you guys know.

rollhard

The problem with using the stock computer is just like simplyscion and I were talking about. 12.5:1 afr is not rich but the ecu will see that is rich and then drop the CEL bomb on you. It will trim so much fuel that it will be dangerous to drive. Trust me on this guys. I log my ECU activity and it was pulling back 45% fuel making me run super lean. After resetting ecu, it would accept it for about 45 min of driving then go right back to freak out mode.

son-of-frimbus

To arizonaturbo:
I did consider the S13/S14 SR20DET T25 turbo and the turbo to a '89-'90 Pontiac Sunbird is very similar, I just couldn't find one that somebody hadn't trashed and I would have to rebuild. The turbo I picked up on Ebay was new for the price of a CHRA, so I snapped it up. It also has an internal wastegate that is set at 7 psi and is adjustable.

To lastlookcustoms:
I agree, speed density systems are easier to use. I still have my '93 Firebird Formula and it has a speed density system. I installed the second 12 psi ATI ProCharger kit on that car (translation: LOTS OF PROBLEMS if you're lucky to be one of the first), but I must admit the speed density caused no problems. In fact alot of the guys that had '94-'95 MAF cars coverted to speed density to avoid the headaches. BTW-I paid $3624 for that kit with a non-eccentric drive pulley (translation: wobble city), a 2 core intercooler bracket to be installed on a 3 core intercooler (they wanted me to design the 3 core bracket and send them an AutoCad drawing of the part that I designed) and I had fuel system problems because ATI's engineer couldn't spell "stoichiometric ratio".

To rollhard:
While I appreciate your opinion, I don't agree that it will cost anywhere close to $2500 to equip my Scion with a turbo. As for the, "If you are doing your own kit, good luck" comment - you're a little too late for wishing me luck, I needed that 12 years ago when I turbocharged my first Fiero with junkyard pieces, a MSD-6 BTM for ignition duties and a ATI FMU. Total cost - $600, but maybe with inflation what it is for the past 12 years --- I don't know.
It seems to me that if you read the beginning of this thread, you would recognize the I am trying to address the problem of the "CEL bomb" by using a voltage dividing circuit on the oxygen sensor. Maybe you should comment on that, or better yet try it - you might find out that it works. Fake out the oxygen sensor - it's worth a try!

As far as your assumption on, "Its not going to matter because the ECU wont be able to recognize the amount of boost going into the motor anyways. Your intake draw wont necessarily by the boost amount," the last time I looked at my fluid mechanics book, air is still a compressible fluid, still obeys the Ideal Gas Law, so for a closed control volume - mass is conserved. Translation into English: Mass in equals mass out, and unless you have nuclear event in between the MAF and the piston that is converting energy into mass - this law still holds. Therefore, if I size my MAF tube by the boost ratio (in my case P1/P is 1.68 for 8.5 psi)for the cross-sectional area AND size the injectors' flow rate by the same amount, the ECU shouldn't know the difference.
210 cc/min. * 1.68 = 361.2 cc/min
MAF Calc. 55mm (stock dia. MAF) Cross-sectional Area = 2375 sqr. mm * 1.68= 3991 sqr. mm, which corresponds to a MAF tube diameter of 71.25 mm. I propose that is how ZPI got their car to run without e-manage and I am suggesting that it can be done with our cars too. The only assumption that could mess me up is that the injectors or ecu might not be able to drop down to the 1.8 millisecond range that I would need for a decent idle without loading up the cylinders, but what the heck, I can experiment. If the ecu is so smart, then let it try do the work.

I'm just trying to make suggestions --- I'm not too interested in getting ripped on, I got a wife for that!

arizonaturbo

iTrader: (1)

I am insterested to see how the Greddy MAP sensor works on our cars.

lastlookcustoms

I am telling you, convert to speed density. That is the best way as it totally removes the MAF and converts to Manifold Absolute Pressure. MAP ECU all the way...

rollhard

Hey man. Sorry it my post came out like a rip. I appologize for that. If you read through various posts on the boards here you can see where some of the comments. come from. As for fooling the ECU on the 02 readings that could work well in conjunction with the air flow corrections. I was just curious, are you going to make your own manifold? Oh and dont worry about the ripping. You arent alone there either. If anyone knows how to flame and get under my skin, its my wife.

TIGRIS

FRIMBUS- good to see somebody who is also interested in theory and understanding of engine dynamics. There are just so many " how much power will this make?" threads etc. It's refreshing to see some real thought go into the process instead of bolting on and crossing fingers. I'll kepp you posted on my turbo install. Thanks

arizonaturbo

iTrader: (1)

I couldn't agree more.... I have been on this board or a couple of years lurking and hoping that someone would start a discussion that I could join in on that could possibly benefit someone to read.

rollhard

I was wondering if any of you considered working out the megasquirt system then. For under $300, it would probably be a good choice compared to tricking the ECU.

Reactor

One way to make MAF sensor think that there is more air going through than it actually is, is to make a small restriction in the tube right at the level where the sensor seats. Very small half peanut size piece of plastic or rubber will make a difference in reading. To prevent O2 sensor signal from compensating a small variable resistor (easyer to dial in) can be used. But... it has been done on older Bosch LH-jetronic to compensate for non-stock hot camshafts and i am not shure how much effort it would take to make this work on todays systems.

son-of-frimbus

To rollhard:

No apology is necessary from you, I have been watching you and SimplyScion field questions and being much more patient with the usual, "I know how to change spark plugs - how much horsepower can I get from my Xb witha power inverter running a Conair 1200 watt hairdrier on high with a "no heat" setting and how much extra horsepower for the fart-can muffler - 300hp you think?" questions. Hey, you earned the right to dish it out in my book. Plus, I think we should start a, "My wife is meaner than your wife" thread.

Actually when I started this thread I had you in mind, because you were the only person who was currently posting their experiments - good or bad you posted them all and you took some heat for doing such, but you had nobody to suggest something to solve the "CEL bomb". When I came up with using the voltage dividing circuit, I thought that you or SimplyScion would interested enough to try it. It really is a simple circuit (translation: even I could wire it up - and I'm not the bright). All you would have to do is get a variable resistor (a potentiometer from Radio Shack) that has approximately the same range in resistance at the oxygen sensor and wire it in parallel to the supply voltage (+5 volt) and the signal return to the computer right at the oxygen sensor. If the oxygen sensor is 30 ohms and the potentiometer is also 30 ohms, then the computer will see 1/2 of the voltage, which could be extermely lean and try to richen up the system. I don't know what the calibration values would be because I don't know what the resistance value is on the wideband O2 that Toyota is using, but with a little experimentation I think it would work. Risk/Benefit analysis: risk- $1.25 for potentiometer, a little time and some solder, benefit - the first kid on the block with no cel AND never having to touch the ecu fuse again. As far as risk to the ECU, you will only be lowering the signal voltage - not increasing it, so no chance of harm. As far as risk to the engine - you can only get a richer ecu, no broken pistons and if anyone knows how to break a piston ring land - it's me (I have got a collection of five in my basement.) Hey if you try it and it doesn't work I'll send you the money for the potentiometer - a money back guarantee! Plus if you act right now, I'll give you a way that you can eliminate turbo lag - 100%. No joke - simple physics that really works. I use this item on my Fiero and I get maximum boost as soon as I hit the accelerator - from a dead stop or on the roll. It's not my idea, I got that one from a guy that wrote an article 20 years ago about how he did it on his Mercur Xr4Ti. The Fiero that has that turbo setup is going from 2.8 liter to 3.4 liter right now (longblock is done, just have to hook up accessories).

I was going to TIG weld up my manifold (I have use of a Miller Tigmate that is owned by my brother in my basement - the welder is in the basement - the brother is in Indianapolis) by using buttweld 304 SS pipe or tube fittings and I was tring to decide whether to buy a cheap 304 SS Chinese header and hack off the head flange for $55 off Ebay or just machine up one down in the basement on my Jet Mill-Drill (cheap Chinese vertical mill). The T-25 flange you can get at Himni-racing out of Florida for $38 (as well as other parts to kludge the system with) which I might purchase, because I hate milling stainless - it takes a ton of time and I'm not getting any younger. I probably will get the butt weld fittings at McMaster Carr - the Walmart of hardware stores. I got some local vendors that I can get schedule 5 (.065 wall) fittings from, (McMaster sells schedule 10, .083 wall) but I haven't called that local vendor in five years, so he might not even be there anymore, but I'll figure that out when I get closer to doing this.

To arizonaturbo: By Greddy MAP do you mean a speed density conversion for our Xb's? If so, I couldn't find anything on their website.

To lastlookcustoms: I couldn't find anything on the web that does MAF to speed density for our cars, but I have been told that I can find stuff even with a flashlight and a guide and GPS, so it's probably me.

I originally considered MegaSquirt and Haltech when I was converting my current Fiero to a turbo setup (this is the second Fiero that I have done this to), but I wanted to have control over the ignition map in addition to the fuel map, so it boiled down to MSD 6 BTM plus MegaSquirt, or Haltech alone with no emissions control or jumping into emulation and burning chips. The prices were all the same at the time I was looking, so I went for door number 3- reprogramming the ecm. I'm not so hot on the idea of hacking up my wiring harness to get in the E-manage (although if you go to Greddy's site they show TBD as a release date for a harness for 1NZFE) and I figured I would try MAF pipe and voltage divider and see how close it is. If I need more control, then Emanage ultimate. Hopefully by the time my warranty expires Greddy would release the plug and pray harness. I really have respect for guys like you and SimplyScion and lastlookcustoms, it takes alot of guts to modify your daily driver and post successes (sp?) and failures, but everyone on this board has great benefit from you guys. I'd better stop there, because I can't breathe out of my nose anymore and I hope that I don't offend anyone that I left out. Keep up the good work.

To TIGRIS: I put 60% of my effort into thought of what I am going to do, 30% of my effort into "bolting and finger crossing" , 8% of my effort into "well why did I just puke the head gasket", or "geez - why am I shooting out oil smoke out now, was that a piston?" and finally 2% for "well this engine is now a boat anchor, do I rebuild or upgrade?"

son-of-frimbus

To Reactor:

I have seen guys write up stuff on changing the venturi area of the sensor, which is what you are talking about, but I believe that causes a non-linear relationship between the voltage of the MAF and the actual airflow. (translation: it causes the voltage in a squared (parabolic) or cubed pattern rather that a 1:1 linear ratio. If you double the airflow with that fix you could get 4 times the voltage or maybe even 8 times the voltage. Using a proportionally sized MAF pipe if you have twice the cross sectional area - you have 1/2 the voltage at any given flow. It's linear 3 times up - 1/3 the voltage. I am simplifying a little too much (there is an offset constant also), but the basic description is still accurate.

The method you suggest for the a "in series" resistor would do almost the same as the voltage dividing circuit, but it is an offset - meaning that you are not scaling the whole function, but you are "shifting" the whole curve downward. This means that the lowest resistance the computer can "see" is whatever you installed in series with the oxygen sensor. If you install a 50 ohm resistor and the oxygen sensor normally goes down to 25, the minimum that you can go down to is 75. With the voltage dividing circuit, if you have the variable resistor dialed to 50 ohms and the oxygen sensor is at 25 the resultant resistance is 16.66 ohms. Lower resistance - lower voltage therefore leaner reading and richer output.

Thanks for the input - I appreciate it!

Simplyscion

good information...where were you last year lol
I actually sold my box back in October and the guy blew it up in March doing a 3 gear standstill burnout on an automatic lol. Lets just say see you later ring lands lol. Man, Im really really miss my box but Im glad we have guys like you, rollhard, lastlook etc who are more into the technical aspect of it. This was all Japanese to me and Lastlook when we were doing this stuff last year...everyone thought we were crazy lol

rollhard

LOL. Is there a way to put some kinda resistor on attitude? I figured 100% attitude+100% resister might come to me as 50% attitude?

Anyhow, I think what you wanted to do would work great. The only concern is how the ECU will react to rich,lean, hot or cold conditions. I think what the other guy meant about the Greddy map conversion is that Greddy offers a map sensor to work in conjunction with the Emanage if the parameters exceed what the ECU can read.

Here is what Greddy says about it:

"If the factory range of map or airflow meter is surpassed, you can incorporate a larger airflow meter or our "Optional GReddy Pressure Sensor and Pressure Harness" to set larger scales of adjustment."

The pressure harnes list price is $35 and the sensor is $120. Its not a huge investment if it works well.

Ok, here is the latest of my CEL bomb and FI drama.

As you know, and I have probably said it many times...What I am trying to do is use the Emanage corrections as least as possible. The ECU needs to see signals from various sources relative to each other. Lets say for example, you have learned in school that 2+2 is 4. But now, I am telling you that 2+3 is supposed to be 4. You might accept it for a brief moment because Im telling you that but it wont make sense to you and your brain will drop the Check Brain light. I can see it in your eyes. Thats just a simple example of it, but there really is more to it. The ECU will see signals from air temp, air flow, water temp, 1st 02, 2nd 02 (btw, I think both of our sensors are narrow band), rpm, throttle position, etc. So, for each % of the TPS, air flow, etc the ECU will know what the AFR should be according to its program. Tricking the 02 signal to the ECU might work, but it might start messing with timing or other stuff.

Wow, all this is giving me a headache. lol.

Onward...............

So even with 310cc injectors, perrin fuel rail, I need to add additional injector duty % on the high end under full throttle. Low throttle and idle, I had to lean it out or else it would stall. What I am trying next is the Walbro GS341 pump. I am going to see if installing that pump will take away the necessity of me having to add fuel at WOT. I will then use the Emanage to trim fuel on the low and midrange TP. I should have this done this week and Ill keep you all updated. Make sense? If this dont work, I will contact Greddy to see how this Pressure sensor works. If it takes absolute manifold pressure and sends the signal to the ECU, that would be awesome. That still gives me one problem though. If I tune it to 12.8AFR, will the ECU still trip out and pull fuel back? We will see.

Ok, just datalogged the ECU while the CEL is on. It continues to try and make corrections. Here is when it "freaks" out. 13 is too rich and 16 is too lean. Go past those readings on our narrowband sensors and the ECU will throw a fit.