Das Beast: My E30 track / street build

[Nader393]

I don't think shifting the cam timing actually makes power, but instead moves it's location (including torque) vs. RPM on the dyno curve.

[JehTehsus]

I need to get down to the dyno soon, but I have a nuke gear and I am planning on spending some time/$$$ adjusting and characterizing different curves.

I have put a few thousand km on the car so far and the gear has not slipped.

[dvallis]

IVT: Involuntary Variable Timing

Amputated the Nuke gear. Confirmed we had slipped to 10 degrees retard. And was probably moving around under hard deceleration. Involuntary Variable Timing. Ugh.



The good news is no valves were bent. All cylinders compression checked to ~210 psi. Compression really improved after break-in.

-Pre break-in: 180 185 204 210 181 180
Post break-in: 210 211 210 210 211 210

Went back to the solid cam gear 0/0 TDC



Re-installed the rad, re-loaded the Das Beast tune, burped the cooling system and drove around some. Ran perfect and felt strong. Ready for a real dyno tune

[dvallis]

Sleeved a block boss from M10 to M8 for a knock sensor next time we dyno



Also adding an aux O2 bung, so the dyno can read AFR

[Digitalwave]

Good progress!

[Lugnuts]

Pre break-in: 180 185 204 210 181 180
Post break-in: 210 211 210 210 211 210

Can you share your timing map once you get done on the dyno?

Asking because I have similar compression numbers and am about to swap to a 284/284 cam because of a knock issue at WOT. Granted mine's an NA build but still curious what you're able to dial in.

[Digitalwave]

Can you share your timing map once you get done on the dyno?

Asking because I have similar compression numbers and am about to swap to a 284/284 cam because of a knock issue at WOT. Granted mine's an NA build but still curious what you're able to dial in.

There's some threads about timing tables in the Alternative Tuning forum. I'd say that NA and FI tables are generally drastically different.

[dvallis]

DigitalWave is right. And I'd go further saying tunes are like snowflakes ... every darn one is different once you start modding things. Would not want someone loading my tune and blowing up an engine.

This article about dynos, knock and MBT is pretty interesting. ForcedFirebird mentioned here that "M20s will knock far after MBT". Another good knock thread here. Some good points I found:

"I have found an n/a m20 stops making power well before it knocks . . ."

"NA ETA with 885 head made the most power with 27° . . ."

"ITB 2.8 we built was left at 30.6° max, it was losing power with anything over 32 . . ."

"NA M20 isnt likely to be knock limited before MBT anyway . . ."

"The M20 head is a semi-hemi, which has a longer burn time - modern pentroof designs are faster, so the M20 needs more advance than say, an M54 . . ."

[digger]

Timing map of boosted at 100kpa and below should be somewhat similar to the same na spec.
lug nuts engine with mahle ms ie dish dome seems to be very knock limited which is a bit unusual given the extent which it is. A nice set of headers/exhaust and decent cam may alleviate somewhat but it doesn’t match those quotes from FF
mine isn’t really knock limited but it uses a mm central dish w/o proper squish action so I expect it to be a slow burner

[dvallis]

Downpipe is back with the extra O2 bung installed for the dyno.



After I installed the downpipe, the downstream O2 was acting up. Jumping all over the place. Bad O2 sensor? I pulled it, set up in a vice and applied a rag with some gas..



Result? O2 is NOT broken. Smoothly sweeps from rich to lean as gas evaporates..

Re-installed the O2 but this time reefed it on very tight. Also cranked down the other bung. That fixed it.

I guess a very small air leak can freak out your O2 sensor.

[matias9]

well, after a while time , i see a lot of progress on this beauty. a lot of good upgrades. and i have some doubts about the MLS HG , how much torque was used , was re torque after a few cycles?. on the other hand the belt with the 0.120 hg doesnt work the "gates racing TB 131 K1" or should be fine?

[spike68]

I've got to admit, I haven't been active on these forums in a long time. Sure i've been lurking to check out anything new (parts, builds, etc) but last week I stumbled upon this build. I can't believe I didn't find it sooner. I suppose it's because I've been recently been coming up with a plan for my e30 asking the very basic questions being asked for Das Beast early on.

After spending my evenings and even time during work reading through this I had to log in and comment on here. Thank you thank you thank you so much for being so well organized and sharing all this information. It's insane how much you know and frankly it's motivating for me to want to understand these details more (especially the electrical work).

I can't wait to keep following your work (sorry to hear about your recent engine troubles).

[dvallis]

O2 Horror

Last Friday we're on the way to the dyno, O2 fails. Bring car back to the shop, do the calibration dance: remove sensor from downpipe, unplug, power on 30 sec, replug, power on in open air, controller calibrates, power off & reinstall. Takes a suspiciously long time to calibrate. Then still fails. OK, bad O2.

Innovate Motorsports LC-2 calls the replacement O2 an 8883. I order one from Amazon next day Saturday. They ship it to Dallas, not Austin. I order ANOTHER one Saturday, they lose it today. Augh!

In exasperation I look local and find a Bosch OEM 17323 O2 that looks suspiciously identical. It is. Exact same plug harness and sensor. Calibrates right away and runs perfect.

Dyno Day tomorrow

[Northern]

I found that my LC-1 from ~2014 and an LC-2 from 2019 both came with an NTK xref'd sensor and not a genuine Bosch. Only noticed when I bought a Bosch 17014 from rockauto as an LSU4.2 sensor replacement for my LC-1 and noticed a bunch of differences.

I don't know if the weird part number is to try to force you to buy through Innovate or what.

[dvallis]

Freaking Finally

Finally finished on the dyno. Now that "variable timing" was eliminated, tuner was able to dial in air fuel ratios until VE table looked right. Cells ranged from 60's to high 90s, as they should. Not my table but looks a lot like this.



We got closed loop boost control working perfectly. Started with P=100, I=0, D=0, then dialed down P until response was underdamped. Next started adding D until critical damping. Finally added some I to null DC offset. Ended up with P=23 I=10 D=8. It tracks perfectly in on 10 PSI at WOT with no boost spikes,



The results below are pretty interesting. Both pulls are at 10 PSI boost. Compared to the 8.5:1 build, power comes on much sooner with the new 10:1 pistons. We pick up 30 ft-lbs and 21 HP at 3000 RPM, then double that by 3800 RPM. Wow. That's way more torque under the curve early. Peak HP is about 6 HP higher, but gets there 500 RPM earlier.

UPDATED GRAPH. TQ/HP WERE REVERSED IN THE LEGEND



The only drawback of this build is we are octane limited on premium 93 pump gas. Tuner had a knock sensor on the block and secondary O2 probe in the downpipe. At full 10 PSI boost, he could only pull single digit timing advance above 5000 RPM. Otherwise it would start pinging, just a bit.

Horsepower is still good on the top end, so I'm curious if the torque difference will even be noticeable at high RPM. If we end up needing more top end torque, an E85 conversion is not out of the question. There are a lot of HEBs nearby that sell it. We'd pick up torque and HP across the board, and lift the last 1000 RPM of the torque curve with more advance.