So I just bit the bullet. Spent 2 days looking at eBay turbo's. Was considering the EMUSA ones cuz you can order the chra separate for the main ones but I'm not sure the replacement chra is balanced, didn't look like it in pic but who knows.

Had a hard time finding one that came with a 5bolt outlet gasket. One that did was too small a/r. Actually, I don't know the size of my current turbo, it was perfect on the b18b1 and would have probably been perfect for the b17, I could get 5psi by 4000rpm if I loaded it or 5psi by 4.5-4.7k if not..I went to look, same m12 casting but there's no markings on it for a/r or trim so..

went with speed daddy - only because shipping to here was the fastest supposedly, otherwise I would have went with something else. Total $170 USD including shipping, just over $200 CAD.

If it lasts as long as the old one (which I only paid less the 50 USD for..was advertised for 50 got it in a package deal so was less) then I guess I'm alright.

I went with the same style, t04e t3/t4 5 bolt internal wg style without internal wg so it will just bolt up.

T04E T3/T4 .63 A/R 57 TRIM TURBO/TURBOCHARGER COMPRESSOR 400+HP BOOST STAGE III
http://www.ebay.com/itm/190627430922

I'll try and measure my old one when it's off. I'm hoping I'm close, I really liked the old turbo, was the perfect eBay turbo for a b18b1 but I have no idea any of its measurements. People say .63 can be too slow, .48 chokes.. wish I knew what I had, would have prob got the same but hey I guess we will see soon

this is my birthday present to myself I guess. ****. I just realized I'm just replacing something broken. Would be more of a present if it wasn't broken, or if I was ordering a real turbo.

hard to say tho. EBay turbo is like $100.. lasts 1 to 2 years.. real turbo is 1000, lasts 5 to 10 years maybe.. especially if you are racing or hard on it. Now it'd be worth rebuilding a real one but that's not cheap either. Hard to say but IF each eBay turbo does over a year then it's more economical.

then again price isn't the only factor. It should be well known that real turbo's generally are more efficient and out perform generics.

also, turbo failure COULD take out your engine. I was lucky - I'm hoping ! I do have some small chunks missing from my compressor WHEEL! Hoping they didn't make it past the IC will have to remove and check.

 

Hondata Intake Manifold Gasket arrived (thanks to member porofek)

Speeds Daddy turbo arrived too. Only shipped out Monday morning which surprises me.

Have a leaky injector man my car reeked like gas last night. My jacket smells like it, my ugly ghost smells like it..


I made a video of the unboxing of the ebay turbo and will upload it soon.

 

Removed old turbo.

autopsy -

This is Doctor Turbo, resident coroner. It is 3:55pm March 17th 2018. Subject is a approximately one and a half to two year old "ebay" turbo, records indicate she had run over 20,000km in her last year of active service.

Preliminary observations indicate the subject's primary shaft came apart. Removing the compressor cover, we find a chewed up wheel and lots of gunky, greasy oil built up. This tends to indicate the seals around the bearings have blown, generally this happens due to large amounts of oil pressure being pushed into the oil intake in excess of the sealing capabilities of the seals.

Removing the exhaust housing from the rotational assembly which definitely shows more of the same. A broken circlip has fallen off the shaft after removal. The turbine wheel rotates in all three axis though wasn't chewed enough to actually come out of the housing.

It is the opinion of Dr. Turbo that oil starvation caused the death of this TC unit, further investigation of the TC's oil feed artery is required.

4.20pm: Dr. Turbo artificially starts the heart examining for oil coming from the artery. None is seen. Restrictor is plugged.

Turbo died due to lack of oil. The restrictor I used became plugged. It was supposed to be a proper turbo restrictor, was only a pin hole though. Drilled it out, now using the restrictor that came with new turbo.

Sounds good, boosts good so far, fixed leaky injector.. now tune.

 

 

Oops.

I didn't tighten my turbo drain enough it backed off haha..

I went to go pull out of a parking lot noticed the oil light and 0psi.. killed it, looked, oops.. put it back on best I could for now. Filled oil.

some coolant is disappearing, maybe. Running a few degrees warm. Could be bubble combined with a tiny leak. Gotta check my rad repair and i had this nylon braid covered rubber line in one spot yhat leaked before. Really hope it's not the HG. Never did a comp test this time around. Didn't notice any coolant going anywhere for a while until recently so..

This turbo is awesome, very little lag..hit boost cut under 4k i think (9psi atm)..

gotta tune more. It's shitty out today. Yay happy birthday *curses weather*

really can't wait until all the bugs are worked out

Turbo didn't line up perfectly with my dp gotta get it on the lift and work it.. unfortunately my neighbors lift broke last night while I was over prepping the turbo while they were lifting one of their Supra's.

otherwise it feels really good.

 

had to drill and tap for the compressor outlet.

 

So I've been thinking about some stuff...

technically, where I live, the average pressure is about 880-920 mbar. So, technically anything over this is boost.

I never noticed eCtune had the ability to adjust it's tuning/maps based on this as well, to show boost when actually over atmospheric. It does have an option to use the PA sensor but it never adjusted the maps, perhaps it changed how the code did something on the ecu but not the maps. If I uncheck auto, enter about 900, it shows boost at over atmospheric. The wierd thing is the field is called "SeaLevel pressure:", which shouldn't be definable as pressure at sea level is constant!

This also basically says that all boost numbers I've referenced previously in this thread are technically about 2psi LOW. My gauge also reads about -1.5 when car is off and isn't calibratible..

This makes me wonder as well, do people commonly reference boost in relation to sea level or current pressure? An N/A engine can't have more than atmospheric (generalisation)....

 

I think Honda calls it an "Atmospheric Pressure Sensor". It will make global adjustments to maps based on it's scaler. It allows cars to drive up and down hills without needing thousands of ECU maps. By taking it off auto and setting the scaler to something else, you're just fudging numbers to see boost. What this is doing to the tune, beyond me, the atmospheric pressure sensor shouldn't need to be tuned for, it just is what it is.

Pressure at sea level can change due to weather, just like any other elevation. Atmospheric pressure is dynamic.

 

the thing is, this setting is under settings - global program settings, has nothing to do with the actual map on the ecu. In fact, it seems like in Parameters, there's no adjustable settings for the PA sensor though I think i can turn it off, at least in ectune. Would be nice to actually adjust compensation but load cells are load cells, xx mbar @ xx RPM automatically accounts anyway.

I know it's just adjusting the software to see actual boost, and this is what I wanted. Wasn't sure at first if the setting affected the map, it shouldn't being where it's located in the software and I've verified now it's not.

never said it did need to be tuned for.

I do think it's interesting half the internet seems to think "boost is only above 14.7."

 

Been having a tough time lately, my girlfriend is in the hospital and I had a client rip me off for over 1k in work, during the slowest part of the year for me.

Have been tuning. Cleaned off my IACV today. Also cleaned the one on the auto, which completely stock has the idle bounce..unfortunately the sedans iacv was clean! So that wasn't the problem there. Was a little dirty on the GSR but now clean.

spent some time adjusting my idle and injector offsets again today. Added some idle timing too. She drops solid to 900 now (target), no bogging around. I still have to crank the iacv duty up a bit to get enough air in and I still have to adjust my cold start a bit as it cranks too long. I need a new battery, hopefully I can exchange it again but I doubt it, maybe get a discount on a new one..had to return my battery charger because it didn't seem upto the task of what I need, tho close, but also needed cash unfortunately.

the new turbo is working great. Boosting hard. I put on a new stock map sensor on the t/b to replace the broken one, also rechecked it and my 3bar but both are definitely fucked. The new honda one seems to better than the stock get one which connects to the manifold from a vac line.

I need to get another 3bar map. I will be hooking up my boost control solenoid in the next few days just still working full boost to do some mapping. I am within 3% of my AFR targets for most cells I've hit and logged recently, there are some far reaching cells I still need to hit bit I'm getting close. These new brakes are good for tuning, nice to be able to hold boost with the brakes for a bit and not have them roast after 5sec.

unfortunately one of my axle nuts came loose (oops wtf..!) Which caused some **** wear on my driver side rotor and a bit on my pads but there's still lots of pad life. Rotor looks kinda shitty already I hope it wears thru quick. I know I torqued it right. I torqued it again to 130ftlbs and staked it good.

OH I also installed my skunk2 camber kit! Relubed my ES IS A bushings and slapped em on. Accounted for the positive camber on the passenger side from the curb. Feels better. Definitely need a laser alignment now tho. Didn't wanna do it before getting the camber kit in. Actually I still kinda wanna resolve my control arm issue too, my passenger one is a stock replacement I threw on, rubber bushings and driver side is the stock with ES bushings. I need a new stock for the pass side and to put on the ES bushings again.. should do that soon, still also have to do the driver side brake of the sedan - it still needs a whole teg spindleknuckle on that side.

so I can't afford megalogviewerhd..i really wanted to buy it but I can't. But what I can do is get a free trial of datagraph, a graphing program and program it to have the same graphs pretty much. Did this last night and now have scatter plots for everything I want.

in the process I found another ectune fail - when you export a csv log from a datalog, you loose a fair bit of the data fields including VTEC activation,secondary maps, etc.

I do believe I have a way around this- tunerview app will save csv datalogs straight from the binary stream. I was thinking of a way to read the .elf log files myself and converting to csv but I don't have the file structure. Although i do have the datalog serial output binary structure, perhaps they are the same.

I think it'll be as simple as using tunerview to save csv's of the binary stream though, which also allows me to datalog without the laptop hooked up for later analysis.

With the graphing app I am using I was able to create a few nice scatter plots, like map vs RPM with afr for color and fuel inj% as point size.

I tried using a few online JavaScript graphers (plot.ly and some raw something one I don't remember) but even on my dual quadcore Xeon Mac pro it was way to slow with only 15k rows. Datagraph is much faster.

I will post some screenshots soon, I only ran the analysis on my desktop and I am at a friend's with only my laptop. The biggest disadvantage is the missing data. The app is awesome though as I can create new calculated data using expressions and functions and such it's like excel for graphing on crack.

I was also able to make histogram style line graphs based on any of the values or counts of values (hits) etc.

It can't do the table mapping thing mega can but I don't really need that, ectune has a lot of that built in. Man, ectune would have been awesome had it been finished. I did find out I can't use my analog inputs on the demon tho, need neptune for that.





you can see the chewed rotor here


say hello to my little friend!

oh yeah

I ******* punched a hole thru my inner fender. I got balljoints with stupid zerk fittings on the top thinking they'd be great and last not knowing you aren't really supposed to replace the upper bj's but the whole UCA,Z BUT whatever pressed them out and new ones in tacked em in for safety and they were ok. It's been mega bumpy (like 4" high ice chunks up and down) lately and I guess I hit a bump and sent the zerk thru my inner fender. Haven't heard any other serious bottoming out or anything and this terrain has been NUTS so I'm not too worried.

I'm seriously tempted to install my swag bae soon too, at least for now for street.

**** I need business to pick up tho.

McDonald's drive thru yesterday..laptop on passenger seat, ectune open..asian kid who works there who wasn't even the one serving me comes running up.. "YOOOO U STREET TUNING THAT TING??" .. I nod.. "Siiiick yooo!!" and then he disappeared before I could give him a card so I could tune the probably stock ls teg he owns or is working at mcdicks to buy.

 

Went to the junkyard today and picked up a couple flat response knock sensors. Am going to wire these into my soundcard so I can listen to and record knock. I may also hook it to the ecu for datalogging but I'm not sure how well a flat sensor will work with the ecu. Wasn't ever planning on using the ECU's knock retard, there is no settings for it and I've heard it sucks on these ECUs and doesn't work well over 5000rpm.

I also got a resonant one from a 7M-GTE, it actually has a resonant frequency of 7.9khz +/-500hz, and the 83mm bore calculates to 6.9khz knock frequency, curious what Toyota was thinking on that one.

Have had a nasty oil leak from my oil pan for a bit, loosing lots of oil and its right on the side where my clutch is so I have oil leaking into my clutch which isn't fun either. Been using about 1 liter per 125km.

also got a exhaust leak at my turbo because the new turbo doesn't align proper which is throwing my wideband off and now I'm rich. And retarded. Lol.

 

Nice out today gonna try and do some work but I'm feeling pretty crappy physically, gonna smoke a joint maybe pop an Advil.

hooked up the knock sensor last night, well one of them. I got two flat response "Bosch" style sensors, one am actual Bosch off a turbo Volvo (still had it's turbo surprisingly, found another Volvo engine on the ground too with turbo intact, couldn't believe how tiny the turbo was for a 2.5L, like 1/2 size of mine)..then I also got a Mitsubishi one off a 07 Malibu.

I tested both sensors, hooked to my scope, by first flicking with fingernail, slapping with palm then rapping with pliers, did it back to back, mostly to compare sensitivity. The response on both was flat but the Mitsubishi had more sensitivity. It was also the one that I was able to get with a connector and pigtail, the other is a cut wire off the sensor (which I could solder onto a plug anyway but..) so I put the Mitsubishi on.

Did some driving tests, I see a lot of noise around just under 4khz, which is about half my calculated detonation fundamental but no peaks at fundimental or any of the harmonics.

The signal looks pretty clean on the scope, gets upto about 0.5v at high RPM. Doesn't seem to pick up a lot of outside noise.

I should note, these sensors have an m8 mount, so I was unable to use the stock mounting hole (which is m10 or m12, I've read m12 but not sure myself) instead I used one of the left over bolt holes from the OEM block breather. This was suggested in a thread on here or somewhere when googling tricks to mount it where I wanted. I don't have an m12 stud, I was thinking of getting threaded inserts, I'm still not 100% sure if it's m10 or m12 and it's in an annoying place to access without having a lift or removing the IM. The guy who suggested it said he could still identify knock fine with this sensor positioning. It's not perfect, I was thinking of moving it up to the hole for the OEM breather hose near the head split if I can, I think that's also m8 and at the same level as the stock knock hole (though I'm aware there's a big opening in the block behind it, it should still pick up what I want)

honestly until I hook up the laptop and induce some slight knock to narrow down the frequency (and LISTEN to the signal) I will leave it where it is. I will be using a professional FireWire soundcard with balanced and unbalanced inputs. Would be nice if the knock sensor had a differential output, lol. I'm not seeing much noise in it though, well under 50mv@idle.


^ Bosch knock sensor first 1/3 is flicking, second 1/3 is slapping, third is rapping

^ Mitsubishi, same test, seems more sensitive.

The extra sensitivity could be good or bad, I haven't tested the difference in the frequency response curves yet. I will be using professional audio software and gear to adjust the signal and band-pass filter it where I need.

here's some driving plots





The last is an almost full boost pull I think even with some ignition fail (need to gap my plugs down a bit, same happened with the b18 at same boost level with the same plugs (new ones) and changing gap helped lots) and fix my exh leak and get my AFRs in check)

you can see see the noise just under 4khz. I can't see this being det for the reasons described above. Also, it appears on fuelcut decel and lightest throttle ie always AND I'm running stock timing in vacuum, so...wonder what is resonating at like 3.8ish khz hmm..my knock fundamental calculates out to 7.1khz supposedly. 3.55khz would be half.

 

Some interesting **** going on in this thread.

 

Thanks!

I will post audio samples as well once I get it hooked up with the laptop. Had to take a nap. Gonna go outside now and drop the oil pan again and make it not leak. Hoping it's not the rear main but it didn't leak before I touched the pan so I doubt it. Gonna try and see what I can do for my exhaust leak at the turbo too, will disconnect the mid and see how far angled down it sits when bolted to the scroll, I'm probably gonna have to make a cut and run across the street and weld it, lol.

 

Ugh double post, so I'll edit and make it useful.

Did some work today after a nap, unfortunately it was a late start so I didn't get everything done.

I had two options to fix my exhaust leak.

Re-cut and weld my downpipe. Well, we reused the flange last time and cutting it again is just gonna be a mess. No one wanted to deal with it not even me, nor my friend across the street that is in the middle of tigging a 4 inch stainless turbo-back with e-cutout. (Yes, 4 inches. On a mk3 supra, 7mgte, big turbo. aem, aem water/meth, I think I posted a pic before of this project, def coming along. I don't think I would have gone 4" personally, it's huge. But I guess, why? Cuz race car! They definitely won't have to worry about their exhaust backpressure being more than boost.)

Option two, grab my old turbine housing and throw it on the new turbo. Crap, do I really wanna take my turbo apart? No.. do I wanna **** with my downpipe? No..

Well, I don't have to **** with the shaft to swap housings so..and both housings are 0.63 a/r.. looked at my old housing..bit chewed on the inside where the wheel hit it. Was a little weary about using it but none of the chewed parts were actually near the wheel.

While the a/r is the same (actually, the old turbo did have the turbine housing a/r stamped into it, but it is inside the inlet so not visible from outside) I had no idea if the turbine exducer and inducer diameters or their trim was the same. If it's bigger, it could hit the housing. If it's smaller, it could not spool as well (just had maybe an aha moment typing that but I'll get to that in a sec)

So.. ripped the new turbo apart...placed the old turbo's housing on the new turbo, gave it a spin, no interference. Yay.

Put it together, buttoned it up, took for drive. My w/b at idle went from "LEAN" (above 20 afr, I think under 22 because at some point it says "AIR") to 14.7 exactly. I guess I guessed the offset perfectly for stoich at least, actually I need to check if the 94 octane I use is E10 or not, lots of gas here is. I think it actually isn't (petro). I think Husky's was, they used to have 90, 92 and 94 e10 everywhere (for same price as other stations 87, 89, 91) then went to 87, 89, 91 everywhere with some having 94. E10 is closer to 14.0 AFR stoich.

Anyway it seems to run fine and boost fine. It does ALMOST seem to be a little more laggy/harder to make boost now but I can't actually verify that yet, gonna have to compare some datalogs. The aha moment I had was realizing that while the turbine wheel didn't hit, it may possibly be a slightly different trim, ie smaller, which would allow some exhaust gas to bypass the turbine and slow response I assume?

I can't measure the old wheel, it's too chewed I think, I don't think any part of the wheel is it's full size anymore, lol.

 

I suppose while I can't Mic both wheels to verify, I can Mic both housings and compare the clearance the wheel has near the exducer and inducer.

I will live with it as-is for now, it felt a little laggier, but that was one test drive and it was late (and I don't like ripping on it so late) so will see.

might measure it sooner than later to verify, I suppose I should have already but it was getting late, I just barely got it together before having to pick my gf up from school. My main concern was that it wasn't going to touch the housing.

It is nice that the exhaust lines up again tho. This housing has less distance between the inlet flange and the outlet? Like less "throat"?



housing from new turbo left, old turbo right. Using right one for now. The shape of the scroll seems slightly different too, but both are stamped as 0.63 a/r.

now that I look at the pics, I think the left one looks like a better shape (for making boost, it's obviously IN better shape). I guess it definitely would affect spool a bit.

will have to drive it for a day or two and see if it's hugely different.

I have a few options though:
1. Leave it like this until I can make a new downpipe, put it back when I make it.
2. Tune my AFRs and then swap the old housing back and deal with the leak until I make a new downpipe.

Will have to decide. The leak wasnt particularly noisy. It did affect my w/b. I don't think it affected it hugely at WOT though and I will verify tomorrow exactly how far off at wot it was. Seems like about 6-7afr at idle different which is huge. At WOT so far, it seems like it's close to before I did all this.

still didn't get around to hooking the knock sensor up to a 1/4" TS cable so no audio yet.

I did do another bit of a pull, afr was only like 13.8 at 5psi (eek) but didn't see any knock signs...

 

Ah crap.

Does Ultra 94 contain ethanol? Does Petro-Canada still offer an ethanol-free gasoline?
Ultra 94 contains ethanol. Due to the fact that various fuel grades are blended at the point of sale, most grades of Petro-Canada fuel may now contain up to 10% ethanol. This represents a change from the previous state, where premium fuel was ethanol-free at Petro-Canada

So, I was right that it USED to not contain ethanol, now it "may"? Blended at the point of sale? Not very clear, sounds like they're blending it in the pump, lol. Are they just saying that to account for the mixing of lower octanes when pumping? Eg. I select 94 octane but there's still some 87 e10 in the hose? I don't see how that could ever mix up to 10%

I wish it was more clear, ie. Yes always 10% or no. "Upto" kinda sucks. I guess one more reason to tune using lambda instead of AFR.

For those who don't know, wideband sensors don't read AFR. They read lambda and multiply that by your fuel type to get AFR.

If my wideband is set to plain gas, 14.7 AFR and I run E10, I will still see 14.7 at stoich, even though stoich for e10 is 14.1 (ie stoich is 1.0 lambda regardless of fuel). I don't think my wideband has a setting for e10, but if it did, it would read 14.1 at stoich using E10.

Really, you can just pretend you are using straight gas and use gas numbers for AFR and be fine..however the problem lies in if you tune for plain gas and then switch.

Say you tuned for 14.7 AFR (1.0 lambda) with plain gas, but then you put in E10, your tune won't be at stoich anymore due to the fuel change, it would be 14.1. Similarly, if you tune for E10 and get it to lambda 1.0 (regardless of whether your wideband is set to e10 and saying 14.1 or set to plain and saying 14.7) and then swap to plain gas, your actual AFR would be 15.2. Another way of saying this is if you are at lambda 1.0 on e10, it will be like 1.1 on gas.

The only thing I don't like about using lambda numbers is the fact that my brain has the right AFR #s drilled into it..and I don't think my tuning software supports it, though I can change my w/b output to show lambda.

I had always assumed (incorrectly) kind of the oppisate, that the sensors measured AFR and then used your fuel type to calculate a lambda number if you wanted it to be shown, e.g., sensor reads 14.7 looks up your fuel type and knows 14.7 is 1.0. But actually the sensor reads 1.0, looks up your fuel type and calculates the AFR # for you.

So you can tune e10 to say 14.7 at stoich as long as it's actually lambda 1.0, or you can set your wideband to say 14.1. Or you can just use lambda.

I also feel like using AFR #s (at least for gas and e10 maybe e85) gives more mental resolution, e.g., it's easier for my brain to interpret the difference between like 14.7 and 13.23 versus 1.0 and 0.90 - like, 2.5 numbers different feels like more of a difference than 0.1 "numbers" different. That, and like I said, AFRs are drilled into my (and probably all of our) head(s), I know I want probably between 11.8 and 12.5 at high boost, without doing the math I'm not sure what lambda that accounts to, I guess 11.8 is 0.8027 and 12.5 is 0.85.. smaller numbers, bigger change.

it may help some to simply multiply by 100 too, just for mental help, e.g. 80.27 is 11.8afr, and 85 is 12.5, there you get 5 numbers different instead of 0.05, mathematically it's all the same, just the scaling might help some visualize it better, especially if used to afr- you might not think 0.05 lambda is a big change but it is while 0.05 AFR change is nothing.

So, I can adjust my wideband to read 14.1 at stoich and then I have to make all the AFR numbers in my tune 0.5 less (targets) which means I'll be targeting around 11.3 max (11.8 for plain).. which kinda makes it harder because again I'm used to seeing gasoline numbers. Or I can leave it set to gasoline and know that 14.7 is stoich even if the engine is running e10 and seeing an actual 14.1.

Really, I just need to train my brain to work with lambda better. I wish I had only ever used lambda, but everywhere you look everyone uses afr. Someone asks, "What's a safe afr for boost?" Someone answers, "Under 13" not "Under 0.89" ..

can be confusing for some I suppose. I will probably leave it set to gasoline and tune to gasoline AFRs but I am going to try and start using lambda more and more. I wish I could show both on my gauge at the same time so I can train my brain to know what afr #s match what lambda (for a particular fuel) so I know where I'm at when it reads 0.8 or 1.1..

 

actually once I get the basic acoustic knock sensor working, I'm considering implementing ion sensing ignition.

I've been reading a few ways to do it, might be able to pull it off with dizzy ign but I also grabbed some k-coils off some random rsx (dunno which k probably base rsx dunno if they'll even fit)

 

So I installed my EGT gauge. Wanted to wait until I had a module to concert k-type to 0-5v for datalogging but meh. Threw the gauge beside my pillar for now lol.

checked my plugs, my spark plug wires had popped off..the wires I got are for non VTEC and a bit short I guess? Weirdly enough some stock-style bosch plugs I got from pnp (def not from VTEC, never see a b vtec in there, even a d VTEC is rare) fit fine as do my old yellow wires off my b18b. NGK FTL. Will have to exchange them. Definitely felt like it picked up some power.

I noticed some fluid coming out of my turbo outlet flange where I don't have a gasket yet, looks mostly oily. Hoping this turbo isn't leaking oil already. Hope it's not coming from motor either. Losing a tiny bit of coolant too it seems like, but not into the oil. Definitely not perfect yet but I'm not too worried. The other b17 block is still on the stand, I'm still planning on getting that sleeved block if buddy still has and yeah, eventually I'll have a fully built fresh one.

I might run a compression test soon to check where I'm at.

I'm getting some turbo spin down whine again, I've read that this can be due to a cracked turbine housing - I just swapped to my old turbo's housing for now (will swap back if/when I make a new downpipe) and this noise happened before on the old turbo..i didn't see any cracks on the housing or any real damage but I dunno. Wierd how a noise like that can come from the turbine side..?

need to adjust my t/b idle screw but I couldn't get it to move when I tried the other day, will have to remove the tb. Currently I have to crank my iacv duty to compensate, the ecu brings it back down anyway to hit target idle but it gives more air and doesn't bog down when settling to idle with the extra air.

hooked up my boost solenoid..finally..only been in the car for over a year and never plugged in..tho that was cuz I had a diff spring in the gate that was way too stiff.. not wanting to spend $$ on a knockoff gate, I got a spring from an r/c car suspension and put it in. Holds about 6-7psi above atmospheric (hits 4-5psi on my gauge) which is perfect.

tested the solenoid, it works. Haven't tested actual boost control or settings yet really. Ran the lines for max boost, 1 from turbo, 2 top, 3 bottom, no tees. Might test other methods. This way I am varying how much goes to top or bottom, can be all top or all bottom or in between. If solenoid fails (assuming its n/c), it will use spring pressure. I'm not 100% sure about this method as it doesn't allow for the pressure to vent any way though it did feel like my solenoid vents itself a bit? Turbosmart shows this method in the w/g manual so it can't be bad? But then, isn't that exactly what a 4port does, varies top / bottom and vents ?

Another method uses a tee..turbo teed to port one and bottom, solenoid 2 vent, 3 is top. This will always have full pressure opening the gate while varying how much fights it closed, even with the solenoid completely open, you get equal pressure at bottom and top ports. This method would give a bit less boost. If solenoid fails (again assuming n/c) you will be using spring pressure.

another method is to use the solenoid to vary pressure to the bottom port only leaving the top port and solenoid2 vented. This is how some factory 2 ports work. If your solenoid fails closed you're screwed. This is your goto if you only have a single port gate and usually what people with internals use. Recommended to use a n/o solenoid here so if it fails, you use spring and don't overboost.

you could also tee the turbo to the top port and use the solenoid to vary the bottom. This could potentially give the most boost probably but if your controller is screwed or the solenoid fails, all boost will be holding the wastegate closed. Not sure I'd ever recommend this setup. As well, vacuum could pull the gate open before you see boost. With the other methods and a solenoid, the solenoid shouldn't start opening until boost so the top port should never see vacuum. Now, I've had only the top port hooked up to my turbo (for testing only) and never saw it get sucked open. In theory my car idles around -10psi, which should be enough vacuum to overcome the spring but I haven't ever seen it open like this, maybe if I had a better gate LOL

I'm just going over plumbing possibilities, not what you should do. The first few are common methods though. This is all with a 3 port. A 4 port adds more possibilities. The method I'm using is similar to a 4 port I guess.

 

https://www.omega.ca/auto/pdf/SimpValvesGuide.pdf

I still really wonder about this connection method without there being a way to vent pressure, but it is recommended by turbosmart..so wtf

 

I soldered my knock sensor onto a 1/4" phone cable to plug into my Profire 610.

I setup my boost controller. It's setup as a diverter valve, diverting boost top or bottom of gate. For some reason, it isn't working normally closed, but with the lines swapped and set to normally open, it works - I gotta double check the plumbing at the solenoid I guess though I must be missing something.

Normally open is 100% on for no boost and 0% for full.
Normally closed is 0% for no boost and 100% for full.

So, normally open, if my target duty cycle is 100% it actually won't activate the solenoid and if it's 0% the actual duty cycle on the wire will be 100% (inverse) I guess. If my solenoid fails like this I can overboost. This way, the solenoid passes pressure from the turbo to the top port when it is off and diverts to bottom port based on duty.

Normally closed, target duty is actual duty and solenoid fails to spring. This way, solenoid should pass pressure to bottom when off and diverts to top when energized based on duty.

Working with it as a diverter like this, I can't see why I wouldn't be able to flip the ports and flip the duty polarity (nc vs no). When I do, I get only wastegate pressure regardless of what duty the solenoid is doing. I am going to have to debug this, perhaps I need to change my port config at the solenoid but again as a diverter it shouldn't matter! I am basically saying send air to port a or b based on duty. Should be no different to say send port to b or a based on duty.

I also for shitsgiggles tried changing nc/No without flipping lines, still spring.

With it normally open ("gm style", I am using your typical MAC solenoid) it works pretty much flawless. It's overboosted a few times, I still need to dial in my duty target table a bit, but it's working good and so is the closed loop part. Fast spool seems to work nice. Low/Hi works nice.

I can also vary boost based on a multiplier vs TPS, so I could say 10% tps is 10% of my target boost. Target is currently set to boost by gear. Unfortunately the only thing ectune doesn't do is boost by gear AND rpm together. There is a low and high target table for each gear or like 11 different RPM columns if using boost by RPM. I think boost by gear makes more sense, I can have lower boost in first and second (although it'd be nice to have both in first at least, then I could maaybe just plant it and go. I need better tires too.) Also, the target can be multiplied by a potentiometer connected to one of the ecu's inputs.

so, Low Boost:
1st gear: 1.99psi*
2nd gear: 3.99psi
3rd gear: 5.05psi
4th gear: 6.00psi
5th gear: 3.05psi

*yes, I'm targeting under spring, while I can't force less than spring it still seems to help as the gate is held closed from like 0-5psi usually for fast spool

High
1st gear: 5.05psi
2nd gear: 8.02psi
3rd gear: 10.02psi
4th gear: 12.46psi
5th gear 6.96psi

spring was measured around 6.0 psi above atmospheric. Atmospheric here is ~894mbar so 6psi is like 1300mbar, at sea level 6psi is like 1400mbar. And people wonder why altitude ***** with cars so much.

my boost gauge isn't calibratible so when it says 6psi it's more like 7.5-8psi actual boost. For example, it shows 4psi on my wastegate spring only while the engine is actually seeing 6psi above atmospheric, 6psi more than it would be getting if it were N/A.

I made some recordings of the knock sensor, with and without dynamic range compression and filtering. I haven't really heard any knock yet. The 4khz sound is clearly valve tick.

unfortunately I can't really drive around with proper headphones on.. I was using good headphones when I first hooked it up but on the road I was using a single **** earbud. Could still hear OK but ear pain FTL.

I can easily stream the sound over Bluetooth to my stereo but I worry about delay. I could also aux it to my deck I guess. I gotta fix my amp wiring too.

I took some videos today that I will post once I edit.. shows the setup and such. I was using logic today but I think cubase may have better analysis plugins, I didn't have a free usb port For my cubase dongle tho.

I tried even loading down boost in 5th and couldn't hear any signs of det. My timing is semi conservative, stock b16a map to boost, my step retard works out to about 1deg/psi but its not linear, like 0 to 3 is 0.08 degrees per psi, 3 to 5 is 0.15, 5 to 7 is 0.25, 7 to 9 is 0.75, 9 to 30 is 1 or so..

well that is what I initally set it to? That's what's saved in the window, seems a bit light..?,
looking at my actual map I guess it's pretty much exactly 1 degree per right now. 30 deg @ 0psi, 27.5 at 2.1psi, 25.5 at 5psi, 20 at 9.4psi, 19.25 at 10.14psi, 16.25 at 10.88psi, 15.25 at 11.52psi, 14.50 at 12.26 psi, 13.5 at 13psi.. so it definitely ramps up after 9.5psi.

I think I probably would have set it to 0.25 degree per from 1 to 3, 0.5 from 3 to 5, 0.75 from 5 to 7, 1 from 7 to 9 and 1.25 from 9 to 30.. actually that sounds exactly right.. so the numbers above in the window I wrote out can't be mine, must be default, that window is an adjustment window not part or the saved tune anyway.

so let's see, if it was where I thought it was, it should be..

30 degrees at 0, 29.25 at 3, 28.25 at 5, 26.75 at 7, 24.75 at 9, 21 at 11. Hm definitely less conservitive than what's on my map but I thought that would be a good starting point. Where my map went from the step regard to this. I don't know. Maybe I ran the step regard twice and doubled it, that could make sense. Now that I have knock ears I can start playing with the timing more and add some in.

I'm really starting to get this tune dialed in. I really need to fix my oil leak though. I've been having some smoke too, I think it's mostly from that. I didn't see any oil/fluid/what ever coming out the turbo/downpipe crack today either. It has been relatively cold here, condensation in engine happens lots. Took forever to get all the condensation out of the motor once it was first in since it was basically thrown in outside in winter.

it's nice when the boost control works.. I can floor it and go..flick a switch if I need more boost and it's there, flick it down and it goes away all in the same pull.

damn glad I got half decent brakes too, helps for tuning LOL

 

I am working on a YouTube video for exp,aiming boost control..

I'm gonna write out my script here to start. If anyone sees any mistakes or inaccuracies let me know.

--

Hello and welcome, today we are going to be looking at how boost control systems work for turbocharged engines.

If you have ever been lucky enough to drive a turbocharged vehicle, you probably know how fun they can be.

This video is going to pass over some of the basics of how turbochargers work because we are focussing on boost control today. If you don't know how a turbocharger works, there are plenty of great videos on YouTube explaining the theory.

Basically though, a turbocharger is an air compressor powered by your engines exhaust gas. Exhaust gas passes over a turbine "fan" causing it to spin, this turbine is connected to the shaft of a compressor which enables it to compress air into your engine.

In order to do so without creating a positive feedback loop and creating so much pressure that your engine explodes, engine designers use different methods of boost control to keep the pressure in check and at levels suitable for the engine it is used on as well as staying within the most efficient range of operation for the turbocharger itself.

There are a few different methods which have been used over the years to control turbocharger boost levels. The main systems are wastegate control, intake air bypass and variable geometry turbochargers. Another closely related and somewhat similar device is a Blow Off Valve or Recirculation Valve but these are generally used for a different purpose which I will touch base on later in the video.

Wastegate control is at and above the most common and is the main topic of this video. Intake air bypass is horribly inefficient and isn't used on any modern engines except sometimes as part of an Anti-lag system in high end race/rally cars. The third type, variable geometry turbochargers are popular on a lot of diesel and truck engines and in my opinion are a great option for boost control unfortunately at this time there aren't really any performance or aftermarket variable geometry turbochargers especially ones meant for smaller displacement (when compared to a big diesel truck) gasoline/spark ignition engines.

Again, the basic premise of any of these systems is controlling boost pressure or limiting the output of the turbo within safe, efficient and reasonable limits.

Let's start with intake air bypass style boost controls while we are at it, we will touch base on blow off / recirculation valves while we are at it. After that, we'll take a look at variable geometry turbochargers and then at then end, we will discuss wastegate boost control including different setup methods, boost controller types and more.

-
Intake Air Bypass

Before I begin, I want to make sure everyone knows that no one actually really uses this type of system for controlling boost anymore, mostly because it is horribly inefficient and because there are better methods. In the earlier days of forced induction systems it was more common and this system makes a bit more sense when coupled with a supercharger instead of a turbocharger. Side note: Instead of being driven by the exhaust gas, a super charger is driven by a belt attached to the engines crank.

An intake air bypass system is exactly what it sounds like. When the engine reaches a desired target boost pressure, a valve is opened and the additional pressure is vented to the atmosphere or back into the compressors inlet.

This method can be horribly inefficient especially for turbochargers. This is because the compressor is working to compress air that you aren't even using. Once you hit your target boost, any extra flow is wasted and the compressor works just as hard.

It isn't as bad of an idea on a supercharger that is directly coupled to the crankshaft of an engine - the supercharger will always be driven with the engine anyway so venting the extra pressure, while technically less efficient, it isn't as inefficient as it would be on a turbine powered compressor (turbocharger).

Some race cars use an intake air bypass system as part of their anti-lag system as this is one situation where you DO want your turbocharger spinning and making boost so that the pressure is there when you open the throttle.

-
Variable Geometry

Another option for boost control is the variable geometry turbocharger. Besides just being boost control, variable turbo's have other advantages.

There are a few different variable turbo designs (Variable Geometry turbo. Variable nozzle turbo, etc) depending on the exact method and company who makes it but the principal is the same on all of them.

They have the ability to alter the airflow to the turbine in a way that makes them responsive at low RPM and still able to breathe at high RPM. Variable turbo's can have much wider efficiency ranges and work across a larger range of engine RPMs while still being incredibly efficient.

Variable turbo's are in a way one or the best TURBOcharger designs. They are commonly used on large diesel engines and trucks but rarely seen on gasoline engines. Some enthusiasts have taken some of these turbo's and adapted them to gasoline engines but even the smallest ones are rather large and it's not a viable option for most smaller displacement engines.

I would venture a guess part of the reason for this is due to their added complexity and control requirements, the average ecu wouldn't know how to control the adjustment or the turbo although with some ingenuity some existing pulse width modulated boost controllers, which brings us to our next topic..

-
wastegate boost control..
to be continued

 

Sold the ys1 b17a1 Transmission to a guy in Ontario for his race car.

 

So ya the ys1 is gone, shipped it off last night.



I hope I packed it good enough. There was a post in here where a guy said he had shipped like 15 b series trannys and posted the weight and a good way to box it up.

I originally planned on building a small crate but said f*ck it because the aforementioned post had a description on how to wrap.

I didn't do exactly that..i wrapped the **** out of it with industrial saran wrap (bought my own that was thicker than the stuff UPS had).. then I put a box inside a bigger box..it was a bit small and split on one end but the bigger box contained that. The top of the bigger box didn't completely want to close either so we taped in some flattened flat rate boxes on top to fill the gap.

The voids between the two boxes were stuffed with more of the wrap bundled up. The outside box was sealed and taped. Then we saran wrapped the outside box with like 20 layers all around. Threw some fragile labels on and taped those down like crazy.

Unless the guy has a good knife it's gonna take him an hour to unwrap it. Lol.

now what do I do with the rest of this stretch WRAP? Wrap my neighbors car?

 

My eBay shifter started rotating again, time to epoxy it again..need to get a better shifter soon, the only slop left in my shifter is from the eBay rod not being wide at the shift linkage connection like the Tegs (it's "supposed" to be for both cars but doesn't include spacers or snything, I spaced it myself and it's almost good but a little left/right wobble still..) otherwise it's the tightest shifter I've had in any car I've owned. Still doesn't compare to my buddy's ITR but I'll get there..! Lol

need to put my center console back in, took it out for paint. Spilled an ice cap on my shifter due to no cup holders cuz they're in the console and I think that killed the epoxy. Note to world and self, ice caps eat epoxy? Wtf

My oil leak has been nasty and I haven't had a ******* minute to fix it. Worst part is that it's getting in my clutch. Which is causing it to slip (and probably eating away at it). It's fine with normal driving and even some boosting but if I shift hard or really go, she slips. Slipped in second the other day and went right to the limiter

It's been smoking too, I assume this is MOSTLY from oil dripping on the exhaust, tho a few times lately I've noticed some oily fluid at my turbo outlet flange. Part of me thinks it was condensation and some oil mostly, I don't think its coolant.

I do have an eBay turbo, I suppose it's possible that it is leaking past the seals (ALREADY?) .. I am using the supplied restrictor. I don't think it is to be honest.

now, I did have to remove my new valve seals because I accidentally put them on before the bottom spring retainers/pads/I forget the proper word atm.. and they were a bitch to get off even though it was right after I put them in, so some might be a little chewed. I tried to avoid chewing the inside part where the valve glides but so,e ,at be a bit out of round or something now. It seems to smoke more at high rpm/vacuum and obviously when the egts are high because of the oil leak. Once the leak is gone it'll be interesting to see if it smokes still and if it does, what is causing it. I'd lean towards the valve seal possibility (which sounds right too with it smoking more in high rpm high vac) but I can't tell with the leak burning off too which is most of not all of the smoaky.

I do have another clutch that came with one of the motors I can maybe throw in if it comes down to it but really it's prob not great, it's used! I don't really know anythign about it, it doesnt lopk toast..i do really like the feel of my clutch and until this it handled everything I threw at it

I mean, the car could totally use a decent clutch with its intended purpose but.. $$$ for a good one. Plus I gotta either pull the engine (if I don't use a lift) or drop the tranny (if i do get a lift)

I'm feeling like utter **** today (yay health problems).. It's windy as ****, looks like it might rain (which would be a nice change from snow but..) but it is probably the warmest it's been this year (like 13° last I checked) so I do wanna try and drop and reseal the pan. When I do that I'll have the clutch cover off and can see if it is disintegrating or how bad it is.

Hoping it's not the rear main - I doubt it. The leak started w/oil pan stuff, would be really odd timing for that to go.

someone send me a clutch? Plz? Lol