VANOS behavior

[varg]

I’m having a weird VANOS problem, luckily I can observe the solenoid duty and cam position directly in logs or real time because I run megasquirt. This is a freshly rebuilt unit that air tested OK. The unit I replaced with it also tested fine but had the same symptom. VANOS engages and disengages (25-26° of advance observed) fine at lower rpm such as when cruising or accelerating slowly, but if I pass ~3,000rpm with it engaged, it will not retard more than a few degrees (less than 5°) until rpm drops under that threshold again, regardless of the solenoid duty i.e. I can command it to advance from 2,000rpm to 4,000rpm and when it passes 4,000 the solenoid duty will drop to zero as it should, but the cam will only retard <5° before I hit redline and will not retard to its base timing until I drop below 3,000rpm again. Normal cycling takes under 1 second each way according to datalogs.

I use the vanos line to feed oil to my turbo via a banjo bolt on the oil filter housing with an AN fitting on it. I don’t see this being an issue since it’s a commonly used part. I also have a shimmed oil pump relief which will produce over 70psi at 5,500rpm, and does around 50psi at 3,000rpm. The cam sprocket spacers are the old style, the one with fewer parts.

Have you seen or had this issue? The web is too polluted with generic VANOS malfunction threads for me to find any concrete evidence of issues like this and their remedies. How does your VANOS behave?

[TobyB]

I'm having a little reading comprehension trouble here:
so, you're able to command full intake advance and get it,
and it'll drop back to baseline as long as you're below 3k.

But if you go over 3k, the cam stays advanced, and won't drop back down even with
oil pressure removed from the actuator?

It sounds like it's mechanically sticking- I've only got experience with the M54, but I have
seen one vanos unit with worn helical teeth.
I've seen many with lots of varnish inside.

t
mostly a gratuitous bump, not a lot of help here...

[varg]

I'm having a little reading comprehension trouble here:
so, you're able to command full intake advance and get it,
and it'll drop back to baseline as long as you're below 3k.

But if you go over 3k, the cam stays advanced, and won't drop back down even with
oil pressure removed from the actuator?

It sounds like it's mechanically sticking- I've only got experience with the M54, but I have
seen one vanos unit with worn helical teeth.
I've seen many with lots of varnish inside.

t
mostly a gratuitous bump, not a lot of help here...

That's the gist of it. I've also had it stick until idle twice and refuse to retard for some time. This only happened with cold oil (I see oil pressures of around 70psi at 2,500rpm with cold oil, coolant up to temp) and makes me think it may be oil pressure related. I did not note any significant wear to the teeth nor dirt while looking at either unit and it wasn't difficult to reinstall. The unit I removed was quite clean, this completely cleaned inside and out rebuilt unit is practically spotless. I've noticed that if you press the spool down into the valve body too far it does hang up on the way back out just a bit, and I'm wondering if due to my higher than stock oil pressure I'm having issues with the spool valve jamming due to a pressure imbalance across it. Doesn't seem right though. It was suggested that maybe the oil I'm using (delvac 15w40) might have something to do with it and that I try a synthetic but that seems unlikely to me. I use this delvac because it's very cheap, perfectly capable of handling short OCIs on a turbocharged engine, and the engine consumes oil. Otherwise I'd put Mobil 1 in it.

[Northern]

Due to limitations with variables vs logging speed on MS41 I don't have any logs with vanos/cam timing logged, and with the state my car is in, I am not going to be able to go find out anytime soon.

[TobyB]

This is totally not the right way to fix it-
what about a stronger spring?
I agree the oil might slow it, but it seems like it ought to eventually drop
to what it's supposed to be, if it's only the oil's viscosity. And that shouldn't affect the valve.
Oh. What if you unplug the valve to disable actuation?

I might be tempted to try to read the oil pressure in vanos cylinder, but that's...
kind of an intrusive prospect on the M52. M54, it's easy enough to drill a hole in the access plug.

t

[varg]

Unplugging it isn’t necessary to disable it, I just turn the output off in my tune. FWIW I did change the oil this past weekend and use synthetic, turned the vanos back on and found that predictably it didn’t fix the issue. I am interested in trying a stronger spool valve spring, it would not be terribly difficult to do, just have to find a spring that is suitable.

[Northern]

I'm wondering if it's a MS config issue? Sticking solenoid (or somehow leaking some oil pressure)?

I also use the same oil feed setup. Haven't noticed an issue (assuming it would throw a code on stock ECU)

[varg]

I am using a different solenoid than I was before the rebuild, it would be unusual for both to have the same issue imo and both bench tested fine. MS output wise, this is a binary setup. The MS turns it off when it should, so it couldn't be a MS issue. I should note that post oil change the rpm it sticks at changed a bit, but it is inconsistent.