engine and drivetrain Found the intake leak in the tii - the warmup regulator

[thehackmechanic]

Going back to many posts in the fall, installation of an air-fuel gauge in my tii revealed that no matter how I adjusted it at idle, it would swing hard lean when revved. I compensated for this by cranking down on the verboten screw to richen the mixture, but the general opinion seemed to be that I still might have a vacuum leak, or have dirty injectors.

I had the injectors cleaned and tested by Jack Fahuna.

I removed the intake manifold and throttle body, cleaned them thoroughly, and changed the gaskets.

I bought a used set of those ugly Top End Performance metal intake plenums that brute-force their way around the plastic plenum's propensity for leaking by using red silicone hoses and 16 hose clamps. Leak NOW, motherf@#$er.

I removed the warmup regulator, put it in boiling water, attached a rubber hose to the connector for auxiliary air that feeds the underside of the intake manifold, and verified that, when removed from the boiling water, I could not suck any air at all through the air hose.

I pulled off the hose from the brake master cylinder and plugged the connection at the manifold to be certain no leaks were coming from there.

I removed the air cleaner and plugged the hose from the valve cover and the little adjunct hose to the side of the throttle body.

I adjusted the warmup regulator to spec and synchronized the pump and throttle body for the hundredth time, started the car, and let it warm up until the WUR was fully extended. I then tweaked the little screw in the tuna can to make the mixture somewhat lean (like 15) so I could easily see it go rich. With an AFM gauge, this is so easy it almost feels like cheating.

Then I sprayed starting fluid at key positions all around the intake manifold while watching the AFM gauge to see if it started to run rich. (Actually, first I shot a quick short blast into the throttle body itself so I could be certain I knew what the response was on the AFM gauge.)

There seemed to be an intake leak under the gasket for plenum #1. Then it seemed to move to #2. I soon realized that the spray from the starting fluid was finding its way into the throttle body.

I connected an old heater hose from my 911 to the throttle body so I could move the source of intake air three feet away from where I was spraying.

I then repeated the test, thoroughly soaking all intake connections with starting fluid.

It turns out that, despite my checking the WUR in boiling water, it is leaking through the WUR. To be certain, I unplugged the air hose from the WUR to the intake manifold and plugged it, and sprayed EVERYWHERE (at least I got my block nice and clean :^) and found no leak.

I carefully measured the extent of the WUR piston, and it's an RCH short of 10mm. I did some searching here and found a post that described a similar problem, and c.d.'s admonition that if you're using a thermostat that's too cool, the coolant getting to the WUR isn't hot enough to bring the piston all the way up. I looked at the thermostat. Sure enough, 75 degrees.

I found an 80 degree thermostat I had kicking around and repeated the test.

Same thing.

So to all the posters who, last fall, replied "sounds like you still have an intake leak are you SURE it's not the WUR," guilty as charged.

But the question is, is this normal, and is it the cause of my lean-running problem? Williamdholmes says in a post "it always bleeds some, even when fully warmed up." So I began wondering how I could leave the valve in play so the engine could warm up but take it out of play once there. It is possible to fully choke off the auxiliary air feed from the WUR to the manifold by squeezing that hose with a set of needle-nosed vice grips, but it's hard to drive the car that way because it interferes with the throttle linkage.

Because I want to know, and because it appeals to my Hack Mechanic sensibilities, I've ordered this solenoid-actuated natural gas valve that I can splice inline with this hose and turn on and off from inside the car.

http://www.ebay.com/itm/New-Electric-Solenoid-38-Brass-Valve-Air-Water-Gas-12v-DC-Volt-NPT-US-SELLER-/121030830113

This valve is NC (normally closed). I would've preferred to find one that's normally open that you'd then need to apply 12V to to close it, but I couldn't. A lot of these small solenoid valves are made for fluid and need a minimum of 3PSI to work. The ones that are made for natural gas, not surprisingly, are normally closed.

I'll let you know how it works out.

But I am thrilled to have finally reached the point where I am convinced through direct experimental evidence that there is no other intake leak other than this.

I would not have nearly this same degree of certainty without the AFM gauge. You can certainly hear a dramatic change in engine revs if you blast starting fluid directly into the throttle body, but the change in revs from the air into the WUR is small. You need the AFM gauge to see it.

[thehackmechanic]

No new ideas under the sun; just found this post from eveofrevolution who put a manually-operated valve in line with the tii's aux air hose.

http://www.bmw2002faq.com/component/option,com_forum/Itemid,0/page,viewtopic/t,328741/start,2/postdays,0/postorder,asc/highlight,/sid,20b18c0aedef6a8aea109e330684fb43/

[ColinK]

Great post Rob and glad you found the source of the leak.

Are you going back to the original plenums or stay with the TEP ones?

Colink

[PatAllen]

fwiw

among all the Tii i worked on, there was always a small leak there. i once disasemble completely such air regulator device and apparently, there is nothing in it that could prevent from leaking off 100%, ie, there is no seals along the moving piston. as it wears out, it leaks more. can potentialy be compensated with the co mixture in the throtle, up to a limit.

i might be completely wrong, but this is what i figured out back then. all thoses cars left my hands like scalded cats.

[thehackmechanic]

Waiting for the valve to arrive, I figured out a safe way to clamp off the aux air hose that doesn't interfere with the linkage, and drove the car.

It didn't make any difference. My car still runs way lean (>15) in a whole variety of conditions unless I crank the verboten screw by three full turns.

(Regarding the plenums, now that I know this isn't the problem, at some point I'll return the original plastic ones).

What's life without mystery?

[bnam]

Rob,

That's what I found as well -- had to turn the verboten screw quite a bit and it still ran lean. Better, but still lean.

I thought it was because I had different cam that breathed differently, but I recently confirmed I have a stock cam.

Could it be that the pump pistons are no longer sealing as well and not injecting enough fuel at high revs?

Also, did you check for an air leak in your exhaust upstream of the sensor -- I've read that that could cause lean readings as well.

Byas

[02fanatic]

Rob,

I'd stick with the original black intake runners. I have one of each, and while I love both designs there's something about those black intake runners....really shows well IMO!

Glad you found your leak....so much effort! I've been fortunate having two very early tii's ('72 and a '73), and I've never had any issues....tii engines are simply amazing!

[achman]

I carefully measured the extent of the WUR piston, and it's an RCH short of 10mm.

FYI RCH=Rat's Crotch Hair

If I am not mistaken

[jhenard]

The Navy had a different definition but that is close enough

[bnam]

Rob,

BTW, there is a way to adjust the regulator back to 10mm if it's off.

Byas

[BillWilliams]

Rob, are you still installing that new valve for testing? I hate to ask this, but you are sure you adjusted the intermediate shaft syncing the fuel and the air?

John W, I think I am reading he did not find his leak or I am misreading Rob's last post?

[thehackmechanic]

Bill, yes, I have set up the system as per the manual and sync'd the pump and the throttle body... many times, with new linkage rods from Jack Fahuna and a new main linkage shaft.

To be clear, though there's no question there is a slight leak through the WUR (and I have not yet installed the solenoid valve, though I probably will), it is not the cause of my lean-running troubles.

I had a long chat Thursday with Dave Redszus (sr) (and to answer your question from several months ago, it was about 15 minutes into the conversation when he talked about "data"). Dave felt that, if I was certain I have no intake leak, it is likely I have a synchronization issue, regardless of the fact that I've set it up according to the manual. He called attention to the fact that the circle that the pin goes into is in fact adjustable (I've read Jim Gerock's long post on this many times). Dave agreed with your earlier comment that, if the system is properly set up, the readings on the AFM gauge should be fairly consistent. Mine is all over the map.

I am going to play around with it by biasing the throttle (rotating it at the attachment point to the central shaft) and by shortening one of the linkage rods so it doesn't open as far as fast.