HugoW

Hi,
I recently turboed my 2004 1.8 VVT engine and for a first attempt and DIY tuning my Speeduino I am well chuffed. However, there is a hesitation when revving up under boost. There can be 934857235403540 causes, I'd like to limit this discussion to 1, which I suspect but I would like to hear other people's experience on this. When I look at a datalog, I see a hesitation in the VVT progress when revving up under boost. The 'desired value' is correct, the signals urging the VVT to advance are there, but still the 'measured value' drops and then corrects back up. My oil feed to the turbo starts at the rear exhaust side of the cylinder head. I took out a gallery plug (1/8 NPT), screwed in a bung and AN4'ed it from there to the 0,4 mm reducer fitting on the turbo. I've seen this done on other cars, for instance on this channel https://www.youtube.com/c/BoostedMiata01 (cannot find the exact clip...)

Please let me know if you have a similar engine, how you did the oil feed to the turbo and whether you recognise the problem / know a solution.
Cheers,

Hugo

Ted75zcar

iTrader: (2)

Food for thought ... Mazda ABSOLUTELY considered sourcing the turbo from that location on the MSM and 323 and elected not to. Of course they had access to real data and engineers as opposed to a claim of "no issues" on the interwebs.

That said, your turbo oil source is unlikely to be the cause of your VVT problems. The MS VVT control algorithm is woefully inadequate to effectively control VVT over all of the operating conditions, I can only assume that speeduino is as bad or worse. VVT requires a minimum oil pressure (hot) to reliably regulate to target with the limited capabilities of the MS control system. The required pressure is somewhat application specific, but generally speaking it is somewhere around 20-30psi depending on head configuration. I would say log high-speed to SD and scatter plot target vs actual to determine where you can actually hit target, but I don't know if speeduino has SD capability. Alternatively you can log oil pressure and configure your VVT target table to not command anything you actually want to hit when oil pressures are not reliably within the range listed above.

Another trick is to not target significantly different advance angles in areas of the table that will be hit during a shift change or throttle stab. So basically, the targets in the columns in the middle of your target table (top to bottom) shouldn't vary by more that a few degrees of advance. This reduces the target sloshing your may experience during these transient events. VVT is a slow process when compared to the load transients from a turbo.

Mr Plow

Really good post by Ted75zcar - completely agree on choosing your fights in the rev range where you may be stabbing the throttle on track and avoiding large swings on VVT as it just won't react. It could potentially move your fueling around as the valve timing will be marginally different depending on how it reacts.

curly

iTrader: (25)

Another issue that you should address, oil galley plugs aren't NPT, they're BSPT. They end up being the exact same thread pitch, so they can thread into each other. However an NPT fitting in a BSPT hole will only catch ~1 thread, and is very unstable, wouldn't want that supporting my high pressure oil line. You can get an NPT tap and tap it out, but obviously you need to be very careful with loose shavings from the tapping process. Alternatively you can track down a BSPT to -4 adapter, or get an NPT die and cut down the threads on your current adapter. Either way, something to look into.

If speeduino had some sort of VVT bias table it might work better, but best thing to do would be to log oil pressure to make sure you have enough for VVT, and make sure you're not even targeting much above 5000rpm, and nothing over ~10 degrees above 4000rpm, there aren't many gains beyond that.

HugoW

Thanks all for the input. I've attached my current tune and a log, feel free to shoot. The best example of the hesitation is seen at around 15.6 seconds, with target at 45 degrees and measured dropping from 32-and-rising to 25 degrees, and then rising again. Please note I deliberately raised all target values by 10 degrees, because I offset my 'cam angle @ 0% duty' by 10 degrees as well. This is because Speeduino does not (or did not, in previous firmwares, I don't know about the current) like negative readings here and with exact settings, it would read -1 or -2 occasionally, causing the controls to stall. So I just shifted everything 10 degrees. Now when it reads 8 (=-2), everything remains happy.

@curly; thanks for the heads-ups on the threads. 1/8 BSPT to AN4 seems hard to find, would 1/8 BSPT (so not tapered) be an improvement / suffice? This is more readily available. Or I see a BSPT male to NPT female thingy, that might do the trick.
I haven't done proper VVT timing to make a table before I bolted on the turbo, I just made a general table from what I can find googling (which sometimes goes up really heigh, like 40 degrees at 4k revs!). I know this is a bad thing, but making a real table required a dyno and I didn't know virtual dyno existed untill recently. So, here's where we are now. I might tone it down a bit as per your advice, see what happens. My turbo is spooling up quite early, so I can compensate any loss with boost

Another thing I noticed in the log is that the duty is maxes out at 45%, which is conform the settings. Maybe that need a bit more headroom, I can raise the max to 50% and see what that does. I also have plans to put in a real oil pressure sensor (not a switch) and hook it up to the Speeduino to log the values.

Current tune: https://drive.google.com/file/d/1xim...ew?usp=sharing
Log: https://drive.google.com/file/d/1wOD...ew?usp=sharing
Me and car making happy noises:
Of course I drove down to Mexico and I don't drive my Speeduino'ed car on public roads.

Cheers,

Hugo