Dialing in a big turbo

[e*clipse]

Hello everyone!

I would like to discuss techniques for dialing in a big turbo upgrade. Any procedures are welcome.

As everyone knows, simply bolting it on and stomping on the throttle can lead to disasterous results.

I've read various forums about the importance of EGT's:
http://forums.tdiclub.com/showthread.php?t=244232&highlight=turbo+dial+in

Watching for the differential pressure between the boost & EMP:
http://forums.tdiclub.com/showthread.php?t=228098

My setup is a VNT-20 (GTA2056) on a 1Z motor w/ RC2. I have an 11mm injection pump w/ stock injectors. I know this is not optimum, but I would like to get it running reliably before moving on to the next (chip/injector) level.

I have VAG-COM, a boost gauge, an EGT gauge that can do peak hold & alarm, and a differential pressure gauge.

Suggestions? Stuff to watch out for?

Thanks a million!

 

[mojogoes]

By remapping the ecu?

 

[Bob_Fout]

I'm with mojo here. Using a VNT20 with an ECU tuned for a VNT-15, there's nothing the owner can do to dial it in.

 

[e*clipse]

From:
http://forums.tdiclub.com/showthread.php?t=228098

mojogoes:
You can only blame yourself to some degree putting a larger unit on and then not adjusting it or changing the programing.............it was probably the quick all ON effect that court it out which put extra stress on the shaft / vanes etc.

I'm trying to avoid doing what happened to TDIsyncro.

As you all know, there's a lot of inter-related stuff happening.

My GTA2056 came off a Sprinter van with an electronic actuator. I made system that could adapt the Vacuum actuator found on the VNT-15.

In addition to actuator length, my setup allows leverage ratio adjustment and has adjustable stops on both the high and low sides.

I'm looking for ways to manage EGT's, the pressure differential between EMP and boost, and possibly optimise spool-up.

 

[Bob_Fout]

Eclipse, those are most, if not all, handled by tuning. You need to get ahold of Jeff for a retune.

There's nothing you can do with RC2 for a VNT-15 (and stock nozzles) that'll help with a VNT-20. You're gonna need MUCH bigger nozzles (fuel to drive the turbo) and a retune.

 

[e*clipse]

So, your saying there's no way to damage this thing by poorly adjusting the actuator mechanism? (OK, reasonably mis-adjusting things) For example, setting the actuator too short? As I understand things, this could restrict the exhaust side causing a high EMP/boost ratio.

I'm not sure if this relates, but I am running an 11mm ALH pump on this thing. As I understand things, the 11mm pump produces a higher injection pressure than the 1Z's pump. This should result in a bit more fueling.

Yes, I could ask Jeff for a re-tune, but I would like it to reasonably well (if not a bit sluggish) before I take the next step. Changing too many things at once is a recipe for problems.

 

[Bob_Fout]

I guess that's possible if it's mis-adjusted, purely from a turbo safety standpoint. Other than that, "dialing it in" isn't possible w/o a retune.

 

[orion2.0]

without bigger nozzles, I dont think you will have enough fuel to create sufficient exhaust flow to power the turbo properly. Do bigger nozzles and a re-tune...then tweek turbo.

 

[mojogoes]

without bigger nozzles, I dont think you will have enough fuel to create sufficient exhaust flow to power the turbo properly. Do bigger nozzles and a re-tune...then tweek turbo.

Agreed!!

 

[e*clipse]

Ok, I'm almost convinced. Perhaps my skull is a little thick...

Question:
By saying "you don't have enough flow to power the turbo"

Are you saying that there is not enough exhaust <heat & flow> to use the turbo to it's capability.

or

There is not enough <heat & flow> to get the turbo to work decently, at least producing enough boost and mass air flow to meet the current requests of the ECU?

also: It appears there is will not be enough heat produced to worry about EGT's at this point.

Could enough differential pressure generated in a poorly adjusted turbo with this setup to cause TDIsyncro's problem?

Thank you for your help. This is making me a LOT less nervous about those first test drives. I was paranoid that I would kill it after all that work.

 

[dieseleux]

In fact... with wide band 02 sensor you see when you put too much air becausse is easy with big turbo and small fueling.
Too much air is too much pressure in intake and also vane are too much close and choke exhaust.
My 11mm with 0.216 are too lean (more than 20:1) at 20psi with my GTB2056VL!
Bigger turbo have slower spool but high effciency in most case, at same pressure than small K03 or GT15, you put more air in cylinder.
15psi are good setting for beginning.
Another point is rpm versus pressure, simple max number is 15psi at 2000rpm, 20psi at 2500rpm and 25psi at 3000rpm and higher.
With this number you never make surge and run your car without trouble before you make custom tunning.



Dieseleux

 

[e*clipse]

Thanks Dieseleux!

Some ballpark #'s and the reasons behind them are VERY helpful.

I see you have an adjustable leverage ratio on your setup.

Have you played the effect of increasing or decreasing the leverage?

I was thinking if the leverage ratio were increased, it may allow faster spool-up, and more quickly open the vanes once the turbo's spooled.

 

[dieseleux]

In my case... i not do test, juste make 1.5x gain, i do my tunning and adjust actuator in ECU.
If you use VNT15 or 17VB chip with bigger turbo, vane open too fast and the result is turbo react slowly.



Dieseleux

 

[e*clipse]

Dieseleux:
If you use VNT15 or 17VB chip with bigger turbo, vane open too fast and the result is turbo react slowly.

Would the lag be bad enough to trigger limp mode, because the requested boost and measured boost don't match?

Perhaps this would work:
Adjust the "vanes shut" stop for a little more open than normal. (This might help avoid chocking the exhaust)
Adjust the actuator so it starts moving a little later than normal. (initial movement at, for example 3-5mm hg. )

Adjust the leverage ration for greater than 1.5, so that once it is moving, it will quickly respond to the ECU's requested value.

Just thinking aloud here...

 

[TDIsyncro]

you have to get a re-tune as a starting point. However, I don't recall what injector you have..you might want to step into the right size to match the turbo before you get the tune. Its one of those things, where once your part way in, you have to go the full course to make it work.

For a lesser known turbo/engine set-up you would follow these steps:
1. using mightyvac hooked up to vnt actuator, ensure rod starts to move at around 2-3"hg, and is fully closed by 17-19" vacuum.
2. install EMP gauge
3. Install new tune for turbo/injectors
4. Log MB11 and while watching EMP gauge and ensure that EMP does not spike to some rediculous number on initial spool (like 70+ psi).
5. Send log to tuner for update tune file
6. or if EMP got to high, adjust stop screw to open vanes more, then try re-log, then send log file to tuner. EMP should be in the range of 1.25 to 1.5x boost
7. install new next rev of tune on ECU and log MB11 again, send to tuner..etc.

Note new EMP, as duty cycle controlled by ECU will also effect EMP. I suspect your fueling will be on the low side for turbo and you will not have EMP problem. You can also get the tuner to delete the low boost limp mode switch for the tuning process.

If you have a more common set-up of turbo/injectors, the packaged tunes are pretty much bang on all ready and the first one will be pretty much the right one., but every engine system will have variances in it that can be helped by tweaking.

EDIT - I should also point out that my injectors nozzles were clogged during the tunning phase and it took a while to figure that out. When the injectors were replaced, the engine ran better right away, and the EMP dropped substantially..ie too much late burning fuel into the exhaust manifold can create some nasty EMP and EGT.

 

[dieseleux]

Would the lag be bad enough to trigger limp mode, because the requested boost and measured boost don't match?

Yes, my chip i use have turn off boost error.

Perhaps this would work:
Adjust the "vanes shut" stop for a little more open than normal. (This might help avoid chocking the exhaust)
Adjust the actuator so it starts moving a little later than normal. (initial movement at, for example 3-5mm hg. )

Adjust the leverage ration for greater than 1.5, so that once it is moving, it will quickly respond to the ECU's requested value.

Just thinking aloud here...

Is important for people make your tune, you match actuator stroke with turbo vane stroke, normal turbo (VNT15, 17VA, 17VB...) have 1/2" stroke vane with 1/2" stroke actuator... my GTB2056VL have 3/4" electric actuator, i just make matching network with pneumatic actuator for good dynamic with ECU.
Dont match turbo with your bad chip... match turbo with actuator and adjust ECU for personality of your turbo.
Your adjustement are good (stop and inital movement) for beginning setup.


Dieseleux

 

[e*clipse]

Thanks Syncro & Dieseluex!

I'll dial it in as well as possible.

The nozzles are stock 184's.

My turbo came w/ an electric actuator - the lever shown earlier was designed to match the vacuum actuator with the stroke of the VNT control lever.

I'm on a beer budget - at least it's good beer. The new injectors/tune won't happen for a bit.

If I can get reliable operation w/ good fuel economy I'll be happy.

 

[e*clipse]

Quick follow up:

After messing around with the actuator link settings, I settled on a compromise that allowed reasonably quick spool up while not pegging the differential pressure sensor too much. It turns out to be basically what TDIsyncro recommended.

With a N75 duty cycle of about 24% at idle, the actual MAP was aproximately 200mPA below the requested MAP.

The differential pressure sensor I'm using (a Murphy switch connected to an alarm buzzer) can handle up to 15 psi differential. How much short term differential pressure is ok?

By rolling onto the throttle slowly, it's possible to keep the differential pressure low. Once at steady state, the requested and actual MAP pressures were identical. N75 duty cycle varied from 39% to 50%.

EGT's were no big deal. They never got above 550F for a short careful
drive.

For some reason, the Glow plug light is blinking after the drive. VAG-COM shows no errors. Any ideas about that?

 

[TDIsyncro]

my N75 values at idle are much lower than that.

Differential pressure....I would think the 35psi range should be fine and it sounds like you are a long way aways from that at this point in time so no worries.

 

[oldpoopie]

For some reason, the Glow plug light is blinking after the drive. VAG-COM shows no errors. Any ideas about that?

Check your brake lights.

 

[e*clipse]

oldpoopie:
Check your brake lights.

That seems to be it. My taillight circuit, (which also connects to the instrument cluster lights) has a blown fuse. I think I have a short due to some wiring changes.

TDIsyncro:
my N75 values at idle are much lower than that.

Do you have any idea what would effect that? Is it possibly the tune on the chip, or the ECU atttempting to adjust for MAP actual? MAP req is higher by about 200mPA while idling.

We took it on a longer test drive with a long hill. This setup is doing MUCH better than the VNT-15. I had problems w/ limp mode w/ the VNT-15, and no problems with this set up.

No smoke.

EGT's during the climb got up to 1000F, at which point I got off the throttle.
Once the turbo's spooled, MAP actual tracks MAP requested very closely.

Max boost pressure during the run was 20psi.

The exhaust doesn't really "whistle" - it "hisses." Is this normal for a VNT-20?

The only problem was (I think) unrealated. There seems to be an oil leak between the head and the block, at the ends of the block. It's a fairly significant leak, resulting in dribbles of oil.

Could this be as easy to fix as re-torquing the head?

 

[TDIsyncro]

yes, I was just pointing out that a tune would probably close the vanes down a bit more to help with spool.

As for the oil leak..are you sure its not coming from the camshaft seals? Also, I am trying to recal the ALH/AHU head design, on PD's the two outer cam bearing caps need to be sealed to head or they will leak..perhaps ALH head could have same issue?

 

[e*clipse]

On the AHU, only the front (closer to the TB) bearing needs to be sealed.

I replaced the seal when I did my timing belt. I'll look closer, but the oil leak does seem to be coming from between the block & the head.

I did try to look between the inner timing belt cover plate & the block to look for leak evidence. It's a little difficult to see w/o removing the timing belt.

It does seem to be flow/pressure related. It only starts leaking when actually driving. Idling & revving the engine doesn't cause any leakage.

Regarding VNT at low rpm:
I tried adjusting the link to shut the vanes more while idling and got an interesting result. It seems you can shut it too much. If I adjust the length of the actuator so that the vanes shut more at idle, I get an increase in the differential pressure (EMP's) and ** not necessarily ** an improvement of spool up time.

BTW, do you get a turbo whistle out of your exhaust, or some other sound. It's hard to describe mine - it's just a lot different than the VNT-15's sound.

 

[Bob_Fout]

Bigger turbos that I've heard go shshshsh and not a whine sound.

 

[oldpoopie]

Just about any leak above the head gasket will look like its coming from the head gasket..

 

[e*clipse]

@Bob Fout - good! I was worried; it was more of a hissing noise - like tires on wet pavement.

@Oldpoopie:
I'll look into the leaks closer tomorrow. I hope you're right.

I found something very interesting in this thread about initial set points for a turbo:

http://forums.tdiclub.com/showthread.php?t=254951&page=15

Post #220