ALH "safe" bost pressures

[BenT]

Hi all, hope I've posted in the right place...

I have a 2002 ALH (on 180,000 miles) and recently acquired a re-mapped ECU. It claims 130 BHP from this engine! I'm a little skeptical about the turbo life expectancy so I thought I'd find out what it does with the turbo. So, I've logged group 11 in VCDS (engine RPM, boost requested, boost actual) and have produced graphs like the following:

X-axis is TIME, in seconds

This was low gear full acceleration. The boost spikes at up to 2600 mbar absolute.

I then tried a 4th gear, 36 to 60 mph WOT test and got the following (see the square orange curve from 1077 to 1085 seconds ish):

The x-axis is TIME, in seconds

This shows the boost spiking at 2200 mbar but then dropping down to a sustained 1900 mbar.

My question is: is this likely to pop the turbo? Thanks in advance!

 

[powermandan]

Have you logged your orginal ECU? Seems like it's trying to make 1 bar of boost. Which I imagine is probably standard. So you would need to check standard log or look at standard file.

If they haven't changed the boost requested, then they haven't N75 map. Which would indicate that this is just how your turbo behaves. It does seem to have a lazy response when trying to reduce boost. But it could be as simple as the VNT is getting gummed up with soot and the ECU is having to shed more vacuum to get the vains to close. Which takes more time for the PID controller to do. Leading to the spike.

Quite within the realms of possibity to get from 110hp - 130hp purely by adding fuel and no extra boost. Although expect a smokier tune

I notice that all of your logs stay under 2500 RPM. You should log all the way to redline really.

My understanding (although I'm not 100% on this, so hopefully someone else will chime In ) is that a boost spike lasting a couple of seconds in the sort of RPM your getting them. Isn't going to damage your turbo. My reasoning behind this is over speeding the shaft will lead to turbo damage. However your turbo will probably be rated to 1 bar boost all the way up to 4500rpm. At 4500RPM at full boost. You would expect your shaft speed to be at its highest because of the volume of gas coming out of the exhaust ports.. So if your at 2250 rpm you would expect the shaft speed to be half of that while at full boost. So my thoughts are that if you are overboosting 0.2bar at 2500rpm the shaft speed on the turbo is still probably within design constraints.

 

[Nevada_TDI]

subtract 1000 for atmospheric pressure, leaving you 1200 millibar convert 1200 mbar to psi and the answer is 17.404 PSI
1900-1000= 13.05 PSI
There are not turbo killing results

 

[Nevada_TDI]

If you have a boost spike lasting longer than three seconds, the ECU will pull fuel.
What it usually looks like is a big puff of black smoke accompanied by lack of power/throttle response.

 

[BenT]

Thanks both for your replies! I will do another WOT across the full rev range and see what happens.

Just a few other observations - I can hear that the turbo's working harder, even though this doesn't seem to be being requested. It's also sounds like it's boosting on the overrun (weird). No smoke at all. And yes, it does look rather laggy in the graphs above. I'll put the old ECU in and re-run too.

I'm very tempted to fit the 11 mm pump I've got sitting in the shed, from when it was an auto... Nut maybe that would be greedy

 

[burpod]

try some throttle plants (not necessarily always WOT!) from other various rpm points , like 2000, 2500, 3000. do a 40%, 60% ,80% pedal "plant" up 1000 rpm or at least 10 seconds. there seems to be quite a bit of lag there for a vnt15 - the lag is almost surely causing the overshoot, and lots of reasons it could be happening... low vacuum, rod length to long, n75 valve not working correctly, sticky vanes, under-fueled, timing not right, n75 not matched to the boost map well, n75 not matched for the pump map... etc etc etc.. without more data, log files, it's a lot of guess-work, and being of an unknown tune, who the heck knows....

 

[Nevada_TDI]

BenT, will you post your Excel logs without using the Malone tool?

 

[BenT]

Thanks all for your responses

Very insightful. Actually, I do need to come clean on a few things... Firstly, the car's running an SDI camshaft (cam locking tool left in + starter motor = D'oh). I couldn't find a TDI cam at a reasonable price so thought I'd try the SDI one. It actually seems to boost more readily at lower RPM than it used to.

Secondly, it has a 10 mm pump and "auto" injectors. Admittedly the power's a little less than with the 11 mm pump, but it runs so much more quietly and with no smoke at all. Which aligns nicely with Burpod's thinking. Anyway, as requested...

The following's with the ORIGINAL, unmolested ECU in place. It's had a flat spot at 2,500 RPM all the while since I did the 02M swap last year and you can see it here! WOT between 295 and 305 seconds, but the N75 decides its work is done at exactly 2,500 revs

This was a 3rd gear, 30 - 60 mph run. Approx 8 seconds.

Next, with the re-mapped ECU. Again, 30 - 60 (so 1,900 to 3,500 RPM) takes a tad under 8 seconds. But it felt much quicker. And no flat spot! The boost spikes are of the same magnitude.

I don't know if I can post the actual full Excel graphs here rather than snapshots?

It certainly feels more keen with the re-mapped ECU and I can hear turbo boost everywhere in the rev range. It's more conservative with the stock ECU. But, the flat out figures seem similar. It runs much nicer with the original ECU - better engine braking, easier starting, softer idle... It definitely feels akin to a hammer mod with the re-mapped one.

Thanks once again!

 

[burpod]

just upload the .csv files somewhere. but sounds like there's mismatched hardware/tune and maybe other problems going on. with 10mm and the tiny auto nozzles, your're underfueled and i'm sure neither of your ecu's are mapped right for it. if you want to get the car running decently quick, you need to at least get the 11mm and at least stock ".184" size nozzles from a manual or ideally something a little bigger and get it mapped right. else you can futz with hammer mod and it may be good enough. and maybe do timing adapation a little bit. 10mm + .157 nozzles must be painfully slow

 

[03TDICommuter]

Interesting graphs, and with your comments about running auto injectors with the 10mm pump, I'm wondering if that's what's up with my Beetle (low on power overall, overshooting). I still still to install the Wuzetem 0.205 that I had DFIS set up, but when I pull out these original injectors, I wonder if I'll find out they too are automatic injectors vs manual that should be in there.

 

[BenT]

Thanks both! Yes, it is painfully slow. 0-60 takes about 16 seconds I think! But, most of the time it's worth it for the lack of smoke and general running manners. Power is better with the stock manual (170 micron, 150P706) injectors, but it smokes a bit and isn't as economical. With the 11 mm pump the power's better, but again smokier and sounds like a tractor! Slightly more economical.

I'll try to upload the CSVs.

In fairness, the car's pretty much had it anyway, so I thought I'd have a bit of fun with it! I've got some 0.184 injectors, and the 11 mm pump... Ooh, the possibilities...

Thanks once again!

 

[Nevada_TDI]

IBW likes the 11mm pump with .184 nozzles and a vnt15 turbo, not a racer by any means, but fun to drive with great FE.