NefMoto

I'm getting ready to install and tune a generic t3/t4 50 trim turbo on my 2003 A4 AMB 1.8t. Yes, I am aware of the concerns of running cheap generic turbos but many have used them with success, and frankly I don't want to dump a ton of cash into this car. I already have all the parts for the build except the downpipe which I will fabricate myself when I take the stock parts out. All the supporting upgrades are in place... 630cc (at 4 bar) EV14 injectors are tuned and fuel trims are pretty much perfect. Also, I have installed a front mount intercooler. For now I'm going to run the stock exhaust with an electronic cutout after the downpipe. Trying to keep it as much of a sleeper as possible. I guess the main question I have is regarding KFLDIMX at this point. I've been doing a lot of research and want to make sure I get this right, but still am not sure how to go about tuning it. My new wastegate is 8lb so very close to the stock 7lb spring. Hoping someone with experience will chime in here and shed some light on big turbo PID tuning. More info on the car and my goals for it are on a website I created which also has a nice intercooler DIY - cgramme.github.io/a4 (http://cgramme.github.io/a4) Thanks in advance for the expert advice.

Take a look at this: http://nefariousmotorsports.com/forum/index.php?topic=12352.0title=

This will at least help you with the PID :)

Think your original rods are up to the task?

I'm curious as well in regard to rods  :P

I've heard every excuse in the book. Had two cars in the past 6 months with holes in the block. It's really sad, because this should be common knowledge at this point.

The information on stock rods I've read has all been pretty consistent. Everything seems to say that anything above 300 whp is when you run into trouble. I'm not trying to make a hp monster, just something fun to drive. Also, Since I am using a journal bearing turbo the boost isn't going to hit until around 4000 rpm which is good considering boost spikes and low rpm boost is a common rod killer. I'm also going to run a manual boost controller in parallel with the N75 to protect from boost spikes.  vwaudiguy, tell me about what you were running when your engine took a dump.

Thanks, KasperH.

.8 lambda, e85, 27° timing full boost, stock rods
 , 3 bar msp sensor, mafless rods 24 psi 367whp 312wheel tq. Going on year 3.
https://m.youtube.com/watch?v=8xzp5p6XODg
http://sendvid.com/bwg1rd1p

I blew up my block with a f21 @ 20psi around 4,5000rpm.  Rod #4 snapped and blew holes in the block.   I took logs that very same day, no knock, no boost spikes everything was running perfectly.  Stock rods with anything near 300ft/lb torque and it's just a matter of time.

My advice, be ready with a new short block to save on downtime.  :)

it comes down to frame size. peak tq at 5,5k is a lot safer then 4000rpm or lower.

This i agree with. I would never recommend it, but i will say you are safer with a bigger slower spooling turbo

Why not use KFLDHBN to cap boost at low rpm to keep the torque down?

This is inspirational  :o  well I'm following along here then, I'm awaiting a GT2860RS and have concerns of grenading but my understanding is low RPM and high boost will be a no-no.

Would running for example a taper of 17-22 PSI from 3000 to 7200 be okay?

I took the approach of limiting torque at lower rpms too.   Worked fine for a couple years, and then one day...BOOOM.

You roll the dice every time you step on the pedal with stock rods.   I enjoy no longer having to pucker my asshole in fear of total destruction every time I stomp on the gas pedal.

I dont get it, rods are cheap. You can do rods for what you save tuning it youtself

Always good to hear success stories with the 1.8t. Any recommendations on the PID tuning with the larger turbo? Have you been running speed density tune ever since upgrading the turbo?

 
I like the idea of not having to worry about blowing it up every time I get on it. What psi were you running at, and what rpm was your peak torque at?


My goals are really pretty mild for this car. Would like to run around 18 psi on 93 oct. I agree though, rods are cheap, I just can't pull my engine right now.

Has anyone done the rods on longitudinal 1.8t in car? Just drop the subframe, remove oil pan and head ?

Check post #2 of this thread
 

Rod blew up @ 20psi at around 4,500rpm, w/ WMI.  Unsure of peak torque numbers, but it was probably around that 300ft/lb range.

Marty's setup will blow too, it's just a matter of time. :)

Seems like the answer to trying to make the stock rods "reliable" is to taper boost. Even if I have to taper from 16-18 psi until new rods go in I'll be happy just to get the turbo in and running. I'll have to read over post #2 thread a few times.

IMO torque kills rods, not high rpm boost.

That is to say, peak torque once the turbo is spooled. For big turbos, this isn't as much an issue, since you never really get a big torque spike like you would with k03s and k04s.

Big turbo induced motor failure is usually overspin, valve train, or bearing failure due to heat/friction/float at high RPMs

Or lean conditions, but if you are monitoring with a wide band, you should be fine.

Or detonation, but if you are using more or less stock like KR, you should be fine.

Long story short, tapering boost will not save your rods.

YMMV, this is a very rough generalization.

Yes absolutely Nyet, torque is the rod killer. Increasing psi does increase torque in relation to timing, fueling, temperature/conditions. I'm assuming the turbo I'm about to be using will hit 18 psi at 4000 rpm so I may try and do some very light tapering and not hit peak boost until 5000+ rpm. I have a pretty sweet setup using the Android Torque app to monitor afr in graph form (time/afr) real time. I'm using stock KR and still am using the stock rpm limiter for now.

Sorry, bit of a miscommunication; generally, I read "tapering" boost as having the boost drop as rpms rise, generally to prevent running out of injector, or allow for more timing up top when using pump gas and/or high compression ratio motor.

Not sure what word I would use to describe delaying peak torque...

And then you can use the program that's on the last page to make perfect boost PID for you.

Sounds magical, gonna read through it tonight.


I can see how it's easily misinterpreted. I definitely appreciate the help though.

I also did a larger small turbo (gt2860r) on stock rods but that is mostly due the block already being at 340.000km and the cost of upgrading the rods will be too large for what is left of this motor. But I can ensure you I did not enjoy the pid calibration with the 70% and up fixed dutycycle on this thing  ::)
To do the rods on a longitudinal, you need to remove the subframe at least. I would prefer to take out the block and take your time, replace bearings and seals etc. or do as I did and source a good spare block and build that one. You can take your time and do things properly. I bought a good block (115.000km) for 600€ and after selling all spare parts it didnt cost me anything anymore.

Since I am on stock rods and really don't want to surpass 18psi for now tuning on a fixed wgdc it isn't going to help me much past 60-70% dc. I guess I'll tune it for what it can take for now and then try and manually guess what it is at a little higher dc for the time being. Never tuned a larger turbo before... what can i expect running the stock PID's and load maps on first start up?

just unplug n75 then :)
its not a rocket science, im prefer to pretty linearize kfldrl, adjust few axes for tighter wg (covered in wiki) and then fine tune kfldimx (but im really want to try SB_GLI pid calibration tool; just no time for that  :-\ ). its much easier than a nervous little turboes

personally i dont get the reason to use stock rods and CR with big turbo setup

the 4 mo has 230,000 miles on it and the head has never been split form the block, ive personally setup and tuned 10+ big turbo setups local and have had a few drive hours to get setup and tuned over the last 3 years or so and none of them have lost a bottom or bent a rod. k04-015, 001, 023, f21, f23 are the rod launchers end of story. the t04e is perfect for a 1.8t its big enough and laggy enough to be rod safe and with a .63 ar housing is perfect for a 1.8t.

The way I see it, it's 5 or 6 flashes, all of which can be prepared ahead of time.   You can go out and get all the logs you need in about a 1/2 hour.

Or, you can try to manually calibrate your PID, which will likely take many more revisions and trial and error, and it still won't be as good as science can make it.  :)

Mine took 1 hour, with me changing the flash after every run, and the boost is spot on with no hunting.
A little overshoot, but WAY better than stock :)

Good to hear. Not sure why some people are adamant about it being a bad idea. Everything Seems to point at < 4500 rpm high torque is the rod killer.



I'm going to have to try it this way. Read over the thread last night and everyone seems to have good results. I just moved though so now I have to find a good road with no 5-0 to do some logs. All the info makes me feel much better about getting it tuned right. I'm just wondering about how to tune for higher wgdc than my engine can handle. I guess I can use a mbc in parallel with the n75 to keep boost down at high rpm and then do some guessing and manually adjust for the tampered areas. 

If my understanding is correct for higher wgdc or control you would probably be working in open loop and adjusting KFLDRL:

https://s4wiki.com/wiki/Tuning#Specifying_requested_boost

You can do a guesstimation for the higher DC%. I wanted to have them all sort of true to reality so I just let her rip. (I did not calibrate for 95%)  If your tune is not 5120 hacked you can probably better step off the throttle over 1.5 bar boost. The above mentioned tool works very neat, I also just exported 8 tunes with different kfldrapp limits and flashed them after each other. Would be even quicker and less dangerous if you do this on a rolling road though (and potentially cheaper, end of 3rd gear is over the speed limit on most roads).

Yeah, but with a stock turbo and stock wastegate.  The PID is already calibrated for you under 12psi.   Change the turbo or any related hardware and it's not as easy to get the PID working at all rpms/boost levels.  Just about anyone can tune WOT boost manually, it's everything in between can be a shitshow.

I just wanted to ad that there is not reason to remove the engine or the sub frame to put rods in a 1.8t,
take the head off, take the front motor mount brakets off and lower the front of the sub frame while jacking the engine up.
Its pretty easy to out rods in any 1.8t.

I just started to disassemble my car to replace with the new turbo, and after draining the oil I found pieces of plastic.  :( I know this doesn't exactly have to do with tuning but I frankly trust this forums knowledge over any other audi/vw forum. This is putting a little damper on my mood and motivation to install the turbo... Here is a picture of what the pieces look like. Hopefully it has something to do with me removing the valve cover recently and installing it back with the same valve cover seal. The large black ring is about 2" diameter. This isn't my actual photo, but the pieces look identical. Thanks if anyone has any ideas of what it could be.

 (http://i1266.photobucket.com/albums/jj538/kutcht1/IMG_0856_zpsvbx6bgmp.jpg)

I'm not certain about the black seal but that beige piece is possibly from the cam chain tentioner ; worth checking as its a known issue.

That is kinda what I am hoping, although I did change my oil about 200mi ago and saw nothing in it. Pretty much the last thing i wanted to see when I was about to run a large turbo. My first thought was rear main seal, but I don't think it's made up of this type of plastic and I don't have an oil leak. Anyway, I continue the turbo install and then look into what this could be before driving it.


(https://scontent-atl3-1.xx.fbcdn.net/v/t34.0-12/24740678_10100268619747757_346416961_n.jpg?oh=826a754a589eea11aaeb30a494b7c837&oe=5A28697F)

Here are the actual pieces I found in my oil today. Some of them were larger but I broke them trying to figure out what they were made of ect...
Everything seemed to be running fine and there aren't any oil leaks to my knowledge. I don't think it's caused any detrimental damage yet but I don't feel ok with driving the car until I find the culprit and replace it. Any other thoughts on what's in this pan of gold?

(https://cgramme.github.io/a4/images/oilCrap1.jpg)
(https://cgramme.github.io/a4/images/oilCrap2.jpg)

It looks like the tensioner pad for the oil pump chain.
Big turbo passat is going strong

https://youtu.be/UHuzl7Cc_Pc

i bet its part of the pcv 90 and the orange thing is dipstick tube.

**Disclaimer** I didn't get a chance to read the entire thread, but has anyone had an engine blow while running water/meth (which was running at the time of failure)?

I ask this as I've ran the equivalent of an F21 and  GTRS elim at 21 PSI both on W/M for 2 years (and ran them HARD) and never suffered a rod failure.

I was also running an AEM E-boost controller at about 85-90 DC, so both turbos were making peak boost as soon as they possibly could.

Yes.   F21, W/M, 20psi... ran it for 3-4 years like this and then one day... kablewy

I have the same experience with chinese k04
Was using it for a year with 1.4 ok
And after some maintenance i forget to connect n75 line to high pipe
Much bars such wow aand bended rod after few pulls

1.8t high torque at low rpm = f'd rods. My engine was running great when I started the new turbo install, but I'm going to wait to drive it until I find the problem. This sucks, but at least I found it before i pushed it to hard. It'll all buff out in the end.

what problem?

pull the oil pan and check the pickup if you havent already. if the oil pump chain tensioner pad is fine send it

Yeah, there's really no way around not dropping the oil pan. From what I've seen it's pain to do though, and I'll probably need to change the oil pump chain tensioner anyway. I inspected more of the pieces from the oil and so far I have found a soft rubber seal like material, hard black plastic with smooth inside curved edge, and a tan-ish hard plastic material with a smaller curved inside radius.

Update on turbo install:
Made oil drain line, fitted new turbo-intercooler pipe, air intake is complete, made new wastegate flange for new location under turbo, and exhaust is spot welded in place. I just need to plug the coolant lines and make a new heat shield/air box along with finish up the welding.

pan is easy, and the parts are what i said they are.
motor mount brackets, front subframe bolts pry subframe down never loosen rear bolts.
pan slides right out, dont have to take the front off nothing.
Enough with excuses just do it.

It'll happen... Just doing one job at a time. Did you have to raise the engine to get the clearance from the subframe? From what I've heard it's a 10+ hour job.

Pulling the pan is easy don’t believe the hype.

Had to weld in another flex pipe section due to an exhaust leak, but all the exhaust is completed now. The air intake is pretty much done, I just need to weld in a 1" fitting for the diverter and make a very well insulated heat sheild/box. I thought the stock oil feed adapter would fit the new turbo, but the threads are to wide. Just ordered a m12x1.25 - m14x1.5 adapter (hopefully it fits).

Custom airbox and turbo blanket on its way.
(https://cgramme.github.io/a4/images/engine2s.jpg)

Time for an update......... I have finally completed the new turbo install. So far I've gone on two small test drives to check if everything is working correctly, and see if there are any leaks. The car feels super laggy but at around 4k rpm it starts boosting and pulling nicely. I let off the gas when it reaches around 1.1 bar but it feels good. Everything seems to check out, so now I'm going to start the tuning process. I'll be tuning as explained in the thread http://nefariousmotorsports.com/forum/index.php?topic=12352.0title= (http://nefariousmotorsports.com/forum/index.php?topic=12352.0title=). I'm having trouble finding an(http://) xdf that includes KFLDRAPP so I can compare with for my 8E0909518AK 1.8t bin file. I'm still a newb at finding definitions... Anyway, until I can tune it with the nice KFLDRL GUI tool I'll just try and tune KFLDRL manually. Here are some pics of the install, and some of the first file I'm going to flash. I'm very open to educated suggestions regarding tuning.

Completed install:
(https://cgramme.github.io/a4/images/turbski1.jpg)

Midpoint:
(https://cgramme.github.io/a4/images/turbski2.jpg)

Here are are some tuning maps I'll start with. I have additionally set KFLDHBN to 2 across the map, which I believe should regulate max boost pressure to 14.5psi.
(https://cgramme.github.io/a4/images/kfldimx.png)
(https://cgramme.github.io/a4/images/kfldrl.png)
(https://cgramme.github.io/a4/images/ldrxn.png)