G25-660 testing on 1.8t.... info here as more "sticky" than facebook

Righto..
those who follow badger5 on facebook will already have seen this but for those who dont, here's some info from first stage testing of the new Garrett G25-660 turbo. Much "hyped" online unit.. GT28 size, GT35 abilities.. (claims from garrett)

So... Test engine is my poor little ibiza racecar.(as usual). Its engine spec in outline is 82.5mm bore, std stroke, 9.3:1 compression, ported smallport (largeport), dimpled cnc head, catcams,supertech etc etc.

Previously running GTX3076Gen2/0.63 and HTA3076/0.63
2bar ~500bhp region

The G25-660/0.72 unit is garrett vband but fitted directly onto my current tial flanged manifold & downpipe, which was a pleasant surprise. Made installation a breeze. Clocked to fit my setup, all the lines fitted. Nice feature of this G25 series is the chra is vband clamped to the hotside so no more bolts to backout of the hotsides! yay! and easy to rotate.. Double yay!

The size of the unit is indeed compact.. GT28 size.. small turbine but different slightly larger turbine than a 28, but way smaller than GT30xx turbine

comparison plots are of different boosts and against the HTA3076 I removed to do this test. boost onset torque onset etc are what I am looking for.

some plots... more to write when I get a mo to add to this.
G25-660 at different boost levels



HTA3076 compared to G25-660


Some "wobbly boost" experienced from the G25.... Hmmm. Not good.


Thoughts are currently that there is a potential wastegate issue here... possibly just sticking (inconsistent onset of boost), but also I believe back pressure related in the oscillation seen when higher boost at higher revs seen...
the ecu is DTA S80 unit, running STC drive by wire... and boost control is fixed duty cycle driven, not closed loop, so the ringing of the boost is not a PID thing.. there is no PID control active. Its a physical thing. Back Pressure is my concern. Bad thing that.

Wastegate is being replaced, it is as old as noah, and has served me well but we did have to weld its valve head back on a couple months ago where it detached itself. Early old large 44mm tial gate. Trying out a new 44mm GFB wastegate this time.. cos I have'nt tried one, and fancy seeing how it is. I will plumb the new gate in dual port tho, as was only ever single port up to now.

So we will resume testing when parts are in and its modified to work again, this time with back pressure measurements measuring at turbo hotsides.

more later...........

We've added a take off on the G25-660 and my HTA3076 so back pressure can be measured when next on the dyno to continue the testing.
Wastegates (old Tial) is plumbed for dual port as a check also, but new GFB one has arrived today as replacement also.


Even fixed the dynos data logger aquisition card.. one died a year or so again, all of them being obsolete, feared the worst but bought a new innovate card and with some rewiring made it work with the old cards so all back up to many channels of logging, so will be able to log boost and back pressure at same time with dyno.

Prawn said:

Nice stuff Bill!

What's the coiled tube off the back pressure take off? Some form of signal smoothing or isolation away from hot gasses?

Can't wait to see the results.

I swapped my tial 44mm gate from single port to dual wkth a mac valve early on in my road mapping and found boost control much better and more consistent.

Click to expand...

hopefully cool it down a little before the poor little silicon tube gets stuck over its end for measuring back pressure

ok, so another weekends work testing this thing..
Saturday pm had a few hours late afternoon, so stuck badgerwagen on the dyno again. Only change being dual ported the old Tial wastegate and now with back pressure measurement. It was inconsistent boost and boost onset again frustratingly.. but the oscillation seemed to have gone. Time to actually replace the old tial for the new GFB gate I bought to try.

Sunday, removed old Tial and fitted new GFB... utter faf, and not a product I will use again.. Not well thought through product fitment wise at all.. Anyhow, it fitted, was also dual port configured and during the process we removed the chra/coldside from its vband for better access, and saw a rub mark on the hotside and turbine wheel. "feck"


Guessing on previous assembly the chra did'nt get pushed in square enough.. but this would explain the weird boost onset and oscillation if catching. Maybe my old tail gates ok.. Too late now the gfb thing is on there.

Resuming testing now with back pressure measurement and stable, consistent boost onset, we could do some more testing.

First back pressure. Seems max boost run ~30psi we see a 1.5:1 pressure ratio: Tolerable level I believe.

Power output and comparison to both the HTA3076 and GTX3076Gen2 I run.

Onset of boost and torque plots below, violet G25-660,yellow hta3076,red GTX3076G2

desertstorm said:

Again thanks to Bill for doing this and freely sharing this information. I am really happy I believed what Garret were saying and bought one a few months ago. That was lucky, better go and buy a lottery ticket . Now with some shiny bits attached I really need to get this on the car.

So how fast does a G25-550 spool on a 1.8T, as I can't really see it spooling that much better . Same size turbine just a 7mm smaller compressor wheel. And how long before Mamba / Kinagawa start knocking out copies. As there isn't anything really fancy or exotic in this except the Mar M turbine.
View attachment 169281

Click to expand...

The hotside/chra being vband also is something else for the mamba's of the world to copy also..